See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/272090088

The Influence of Selected Indicators on the Risk Behavior of the Philippine Stock
Exchange Index

Thesis · January 2008

DOI: 10.13140/2.1.1971.2809

CITATIONS READS
3 7,960

1 author:

Jeaneth Balaba
Court of Appeals
10 PUBLICATIONS   3 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Program Course Offering View project

All content following this page was uploaded by Jeaneth Balaba on 10 February 2015.

The user has requested enhancement of the downloaded file.

THE INFLUENCE OF SELECTED INDUSTRIES ON THE
RISK BEHAVIOR OF THE PHILIPPINE STOCK
EXCHANGE COMPOSITE INDEX

A Master’s Thesis Presented to the Faculty of the

College of Business Administration

Graduate Studies of Business

De La Salle University-Dasmariñas

In Partial Fulfillment of the Requirements

for the Degree Master of Business Administration

by

Jeaneth Michelle L. Balaba

26 January 2008
ii
 iii

BIOGRAPHICAL SKETCH

Jeaneth Michelle L. Balaba is M.B.A. candidate under the Techno-

MBA Program of the Graduate Studies of Business of the College of

Business Administration at the De La Salle University-Dasmariñas.

She earned her degree in Bachelor of Science in Economics at the

University of the Philippines in Diliman, Quezon City, where she graduated

cum laude in 1988.

She is presently Records Officer III, Section Chief of the Terminated

Files Section of the General Services Department, at the Government

Service Insurance System. She has been on special assignment at the

GSIS Bids and Awards Committee Secretariat since March 2006, where

she prepares the minutes of the meetings and assists in the procurement

process. She reports to VP-GSIS Law Office and Head of the GBAC

Secretariat.

She recently graduated among 35 GSIS scholars in the GSIS

Management Development Program, which culminated in September 2007.

She is a Career Executive Officer (CEO) Candidate in the Career Executive

Service Board.

She was born and raised in Cagayan de Oro City. She lives with her

mother and sisters in Soldiers Hills 4, Molino, Bacoor, Cavite.

 iv

ACKNOWLEDGMENT

Thank you goes to the following people for their invaluable help, support

and contribution to the preparation of this masteral thesis:

My thesis adviser, Ms. Marie Dy, for her patience and attention to my

writing and research effort; my statistics adviser, Dr. Dante Garcia of the

U.S.T. Graduate School;

My informal advisers, Malou Acosta, Richie Cruz-Angnged, Annette Go

and Maiton Fernandez-Zamora for the encouragement in all my

undertakings and for listening to my ramblings;

My new-found friend at work and fellow Management Trainee, Mary Alexis

Rose Bactad-Donato, for volunteering to read through my thesis and for

being very patient and inspiring me to write the closing chapter and for

believing in me;

My group at work, Lea Carreon, Anna Abu-Ngo and Vina Carretero, for

keeping me in good company and giving me sound advice all the time; my

oldest group network at the office, Nelia Valeriano and Rowena Fernandez,
 v

for the writing tips and the coping mechanism advice for this my turn at

MBA thesis-writing; my colleagues at the General Services Department who

witnessed my struggle through MBA school; my colleagues at the GBAC

Secretariat: Tina Reyes, Pia Bugayong, Agnes Santos, Edmon Bautista,

Joedel Alora, Ronald Manila, Mariz Aquino, Angel Velasco, Jed Dañosos

and Gelanie Mariano for being supportive and giving me breathing space

when I became self-absorbed; my colleagues at the GSIS Management

Development Program for the pep talk and small talk and sharing of

learning experiences which have lightened the load of thesis writing,

especially Ms. Valerie Marquez for scanning through the material, and my

MDP buddy Ms. Belen James for her willingness to scan through the

material;

My bosses at the GSIS, Mr. Ted Varela, Ms. Prescy Bernarte, Atty. Nora

Saludares, Ms. Shirley Florentino and Atty. Joy Legaspi, the GBAC

members especially Mr. Dan Cuasay and Mr. Barry Poliquit, for the

confidence and trust in my capacity to perform my job functions even while I

am taking my MBA;
 vi

My MBA classmates and colleagues at the DLSU-D Graduate Studies of

Business for the camaraderie, you know who you are; my mentors at the

DLSU-D, my professors and Ms. Les Molina, GSB Program Director, for the

faith in our undertaking and loads of helpful advice; Mr. Jorge Baula of

DLSU-D GSB for sharing ideas and thesis paper; Dr. Maribec Campos for

helping me secure the needed signatures for the approval sheet;

Mr. Jay Peñaflor and Mr. Mike Moraña of the Philippine Stock Exchange

Market Education Group and Library;

My Business Research Technique professor, Dr. Ruben Nayve, Jr., for

encouraging this topic and for seeing this thesis through its birth pains;

My GSB and DLSU-D family for lending invaluable time and library

resources;

My family and other good friends, for the support, as ever;

To the Lord Almighty for sustaining life and purpose into this effort.
 vii

ABSTRACT

Title: The Influence of Selected Industries on the Risk Behavior of the

Philippine Stock Exchange Composite Index

Researcher : Jeaneth Michelle L. Balaba

Adviser: Professor Maria Luisa G. Dy

Year Completed: 2008

Type of Document: Master’s Thesis

No. of Pages: 116

The study determined the influence of the financial, industrial, oil,

and property sectors to the volatility of the Philippine stock market using

time series data from January 1997 to December 2005. The interplay of risk

relationships was considered in the light of the political environments and

macroeconomic shocks that characterized the period under study.

This paper utilized multiple regression analysis using least squares

estimation in testing the hypothesis that the daily trading activities of the

Philippine stock market, as measured by the Phisix (as the PSEi was

known in this period) representing 30 best-performing stocks, is influenced

by the behavior of the industry sector stock price indices. The stock price
 viii

index in five (5) sectors was analyzed against the PSE composite index

(Phisix). A total of 2349 (days of trading index) samples were purposively

collected from the Philippine Stock Exchange.

Empirical results show that the finance, industry, and property

sectors influenced positively the growth in the Phisix. Mining is not

statistically significant in the model and was dropped from the prediction

model. The oil sector pulled down the Phisix value within the period covered

in the study. The results are significant at 5% level. The inclusion of political

administration was considered after the Phisix model showed structural

instability at 10% level.

Keywords: Prais-Winsten, multiple regression, Philippines Stock

Exchange Composite Index, Phisix, dummy variable,

CAPM
 ix

TABLE OF CONTENTS

Page No.

Title Page i
Approval Sheet ii
Biographical Sketch iii
Acknowledgment iv
Abstract vii
Table of Contents ix
List of Figures xi
List of Tables xii

Chapter I Introduction 1
1.1 Background of the Study 5
1.2 Statement of the Problem 10
1.3 Objectives of the Study 11
1.4 Hypothesis 12
1.5 Assumptions 13
1.6 Significance of the Study 14
1.7 Scope and Limitation 15
1.8 Definition of Terms 18

Chapter II Review of Related Literature 23

2.1 Review 23
2.2 Synthesis 30

Chapter III Conceptual/Theoretical/Operational Framework 32

3.1 Conceptual/Operational Framework 33
3.2 Theoretical Framework 33
 x

TABLE OF CONTENTS

Page No.

Chapter IV Methodology 36
4.1 Research Design 36
4.2 Setting 37
4.3 Participants 37
4.4 Research Instrument 41
4.5 Procedure for Data Gathering 41
4.6 Derivation of Raw Data 42
4.7 Model Specification 44
4.8 Statistical Tests 47
1. Test for Individual Parameters 47
2. Test for Goodness-of-Fit 48
4.9 Data Analysis 49

Chapter V Results and Discussion 50

Chapter VI Conclusions and Recommendations 67

Bibliography 76

Annexes 80

A Summary Matrix 80
B Phisix Performance Across Political Administrations 81
C Regression Results for PSE Index (Phisix) 82
D Iteration of Estimation Model 84
E Sectoral Composition, % of Sector Companies in Phisix 106

Curriculum Vitae 108

 xi

LIST OF FIGURES

FIGURE TITLE Page

1 Author’s Diagrammatic Concept of the Types of Risks 3

2 Conceptual and Operational Framework 33

3 Predictive Power of the Phisix Model 60

4 Actual, Predicted and Residual of the Phisix Model 60

5 Effect of GMA Administration to the Phisix 66

6 Application of the Phisix Model on Day 2349 66

 xii

LIST OF TABLES

TABLE TITLE Page

1 Industry Sectors in the Philippine Stock Exchange 34

(As of 2007)

2 Industry Sectors in the Philippine Stock Exchange 35

(As of 2004)

3 Sectoral Index Composition (1997-2005), PSE 38

4 Companies Comprising the Philippine Stock Exchange 39

Composite Index (PSEi) – As of 16 November 2007

5 Companies Comprising the Philippine Stock Exchange 40

Composite Index (Phisix) – As of 1 December 2005

6 Regression Results for the Phisix 55

7 Predictive Power of the Estimated Phisix Model 59

8 Effect of GMA Administration to the Phisix 65

 1

CHAPTER 1

Introduction

A stock market index is a listing of stocks and a statistic reflecting the

composite value of its components. It is a tool that represents the

characteristics of its component stocks which bear some commonality such

as trading on the same stock market exchange, belonging to the same

industry, or having similar market capitalizations. A broad-base index

represents the performance of a whole stock market — and by proxy,

reflects investor sentiment on the state of the economy. The most regularly

quoted market indexes are broad-base indexes comprised of the stocks of

large companies listed on a nation's largest stock exchanges, such as the

American Dow Jones Industrial Average and S&P 500 Index, the British

FTSE 100, the French CAC 40, the German DAX, the Japanese Nikkei 225,

the Indian Sensex and the Hong Kong Hang Seng Index. In the Philippines,

there is the PSEi, or the Philippine Stock Exchange Composite Index,

formerly known as the Phisix.

The Philippine Stock Exchange (PSE) was incorporated in July 1992

as a non-stock corporation that will provide and maintain a convenient and

suitable market for the exchange, purchase, and sale of all types of
 2

securities and other instruments. In August 2001, the PSE became a stock

corporation.

The PSE traces its roots from the country’s two former bourses: the

Manila Stock Exchange (MSE) and the Makati Stock Exchange (MkSE).

Founded in March 1927, the MSE was the first stock exchange in the

Philippines and one of the oldest in the Far East. Originally housed in

downtown Manila, the MSE moved to Pasig City in 1992. The MkSE, on the

other hand, was established in May 1963 and became the second bourse to

operate in the country. It was based in Makati City.

The two stock exchanges were unified into the Philippine Stock

Exchange in 1994.

In 1996, the PSE Composite Index became a part of the Hang Seng

Asia Index. The Hang Seng reflects the overall performance of emerging

bourses in Asia. Other members of the Hang Seng Asia Index are: Hong

Kong’s Hang Seng Index, Indonesia’s JSX Composite Stock Price Index,

South Korea’s Composite Stock Price Index, Malaysia’s KLSE Composite

Index, Singapore’s SES All-Singapore Index, Thailand’s SET Index, and

Taiwan’s TSE Capitalization Weighted Stock Index.

 3

This paper is essentially a study of risk. It utilizes a proven

methodology in economics and an established framework in finance to draw

out astute observations about Philippine business and industry, and

investing in it.

Risk is a much-dissected concept. In finance, risks are generally

classified into two: systematic risk, or market risks, and unsystematic risks,

or firm-specific risks. These relationships are illustrated below (the dotted

line denotes the risk type typically faced by a corporation, that is, the

combined risk or total risk):

Figure 1. Author’s Diagrammatic Concept of the Types of Risks

Risk

Systematic Unsystematic

Market/Relevant/ Diversifiable/
Non-diversifiable Firm-specific

Combined
Risk

Interest risk Business risk

Inflation risk Financial risk
Maturity risk Total risk Default risk
Liquidity risk Corporate risk
Exchange rate risk
Equity risk
Equity index risk
Political risk
 4

Market risks are risks from movement in factors that affect the

economy as a whole such as interest rate, unemployment, foreign

exchange. Unlike firm-specific risks, market risks cannot be diversified

away. All investments may be affected by market risks, or systematic risks,

but often not to the same degree. For example, sales of products such as

housing and vehicles may suffer more than sales of basic necessities in the

event interest rates rise dramatically.

On a global scale, macroeconomic risks include exchange rate

fluctuations, inflation, interest rates, default risk, and industrial production.

Firm-specific risks are caused by actions that are specific to the

company such as management decisions and labor characteristics. It is

unsystematic because the impact of firm-specific risks is not felt throughout

the economic system. In a given portfolio, the effect of firm-specific risks of

one company may offset the effect of firm-specific risks in another

company. Thus, firm-specific risks are diversifiable or negated by other firm-

specific risks when a combination of stocks is held.

Market risks and firm-specific risks may combine to form the

combined risk which becomes the total risk or corporate risk faced by a firm

or asset. In these instances, a typical firm would face both systematic and

unsystematic risks.
 5

In the pricing of an equity asset, such as a stock, a number of factors

may explain the volatility of the prices.

This study looks into the market risk behavior of the Philippine Stock

Exchange with respect to how singular business sectors behave. It attempts

to bring into common-term understanding the interplay of the market risk

behavior as depicted in the workings of the Philippine Stock Exchange, in

particular, and Philippine business and industry in general.

1.1 Background of the Study

For the purpose of this study, the focus is on market risk within a

portfolio of equity stocks or securities traded at the Philippine Stock

Exchange.

The influence of the overall performance of the industry stocks is

measured against the general behavior of the Philippine stock market as

measured by the Philippine Stock Exchange Composite Index or the Phisix

as it was known prior to the renaming of the index to PSEi in 2006. The

underlying theory for this study is the Capital Asset Pricing Model (CAPM)

which supports the proposition that the riskier the asset or stock with

respect to the return on the market portfolio, the higher the returns on the

stock. This is expressed in the algebraic equation which may simply be

denoted by “y=a+bx” where b, or beta, is the measure of the risk

 6

relationship between the return on asset y and the return on the market, x,

and a is the risk-free rate.

The risk of a portfolio comprises systematic risk and firm-specific

risk. Systematic risk refers to the risk common to all securities, that is,

market risk. Market risk is the risk that the value of an investment will

decrease due to moves in market factors.

Firm-specific risk is the risk associated with individual assets. Firm-

specific risk can be diversified away (or spread out) while systematic risk

(within one market) cannot. Depending on the market, a portfolio of

approximately 15 (or more) well-selected shares might be sufficiently

diversified to leave the portfolio exposed to systematic risk only. The beta

coefficient invariably refers to the measure for systematic or market risk, the

risk that portfolio distribution of assets does not do away with.

In analyzing the risk behavior of the Philippine stock market with

respect to industry sector stock price performance, the study is concerned

primarily with the beta coefficient as determinant of the market or

systematic risk, which is broadly the uncertainty that all financial assets (in

this case, equity assets, or stock prices) are exposed to and which cannot

be diversified away.

There is a concern whether the Phisix is the valid measure of

riskiness. The basic tenet in finance proposes that the market risk is
 7

measured by the risk-free interest rate, which in this case is the T-Bill

interest rate. Because the stock market is a riskier investment vehicle than

T-Bills, there is an expectation that the Phisix yields a higher rate than the

T-Bill.

Therefore, while the T-Bill rate reflects the risk-free interest rate, the

Phisix estimates the risk premium for investing in the stock market. The

Phisix does not equate to the risk-free interest rate as measured by the T-

Bill rate. Rather, it is expectedly valued at an added premium to the risk-

free interest rate. In this study, the impact of individual stock market sectors

are considered as variables determining the value of the Phisix. As far as

this thesis is concerned, the interplay of the risk-free rate while looking at

the risk behavior impact of industry stocks was not considered.

The four standard market risk factors are: (1) Equity risk or the risk

that stock prices will change; (2) Interest rate risk or the risk that interest

rates will change; (3) Currency risk or the risk that foreign exchange rates

will change; (4) Commodity risk or the risk that commodity prices (i.e.

grains, metals, etc.) will change. Sometimes, a fifth risk factor is also

considered and this is the equity index risk, or the risk that stock or other

index prices will change (www.en.wikipedia.org/wiki/Market_risk).

For the purpose of this paper, the fifth risk factor may be the most

relevant that it will be the aggregate movement of stock prices in specific

 8

industry sectors that will be measured against the stock market indicator,

which is the Phisix.

Beta, or the beta coefficient in the linear algebraic expression,

y=a+bx, is a measure of a security’s or portfolio’s volatility or systematic risk

in comparison to the market as a whole.

Strictly, beta is a risk measure that arises from the relationship

between the return on a stock market index and the return of a particular

stock on the market. In this paper, the concept of beta is adapted to

estimate the relationship between industry indices with the stock market

price behavior over a period of time.

Beta is calculated using regression analysis. Beta is the tendency of

a security’s returns to respond to swings in the market.

For the major part of the literature, the beta coefficient is viewed as

the measure of such market risk between country market and the world

market (Li and Lin, 2003), or the world market return and the risk factors

(Ferson and Harvey, 1993), or stock returns and market portfolio (Odabasi,

2000). Shown algebraically where two variables (x and y) are measured

against each other, one the dependent and the other the independent

variable, beta is the coefficient of the dependent variable. It measures the

relationship of the independent variable, x, to the dependent variable, y.

 9

The beta coefficient computed from the study of these relationships

between variables is the critical measure of risk and riskiness of the stock

market to the volatility of Philippine industry stocks.

This study takes off from the CAPM by arguing that the beta

coefficient measures the responsiveness of the stock market as measured

by its proxy index, the Philippine Stock Exchange Composite Index, or the

Phisix, to movements in the industry-level stocks, as measured by the

particular industry indices discussed here.

Elton and Gruber (1994) proposed that the risk of a firm should be

determined by some combination of the firm’s fundamentals and the market

characteristics of the firm’s stocks. By looking at the equity index risk, it is

argued that the effects of firm-specific risks on our portfolio would already

be averaged-out inasmuch as the index should diversify away those

inherent risks that individual firms in the portfolio are subjected to. In

essence, this would make way for the argument that given efficient and

effective management of Filipino firms, the higher the market risks in

Filipino business and industry associated with the growth of the Philippine

stock market, the higher the returns on the stock market. That is, there is a

direct relationship between business management of the market risk in

industries and the growth in the Philippine stock market.

 10

To summarize, this study looks into the concept of risk as it applies

to financial markets when it illustrates how the Philippine stock market

behaves with respect to the movement of stock prices of industry sectors.

This study highlights the importance of the beta coefficient in

measuring the risk relationship of industry sectors to the Philippine stock

market as a whole. By looking at the risk behavior of an index rather than a

particular stock or asset, it veers away from the CAPM model which holds

that the return on a security depends on the risk relationship with the

returns on the market portfolio.

Finally, it validates the use of the Philippine Stock Exchange

Composite Index or the Phisix, which represents the thirty best-performing

Philippine stocks, as a true indicator of the health of the Philippine stock

market.

1.2 Statement of the Problem

The study determined the impact that specific business industry

sectors have on the behavior of the Philippine stock market in general. It

looked into the relationship of sectors to the behavior of the Philippine stock

market as measured by the Phisix. It is argued that the strengthening of the

industry sectors in all of the industry sectors makes the Phisix stronger

even as it represents only the 30 best-performing Philippine stocks.

 11

The study looked into whether industry indices move together with

the Phisix and whether there is a symbiotic rather than competitive

relationship between companies that determine the value of the Phisix and

those that do not enter the Phisix equation.

The research focused on the problem of measuring the influence of

Filipino business and industry on the growth of the Philippine stock market

as they manage the market risk in their respective areas of business

expertise. However, all these interplay of risk relationships were considered

in the light of the political environments and macroeconomic shocks that

characterized the period under study.

1.3 Objectives of the Study

Specifically, the study sought to:

1. Determine the critical industry sectors that influence the

movement of the Philippine stock market;

2. Evaluate the behavior of these industry indices along with the

movement or changes in the Philippine stock market indicator;

 12

3. Assess the impact of the political environment (referring to the

three political administrations during the length of the period of

the study) in relation to the model that is developed.

1.4 Hypothesis

This thesis proposes that the stock price behavior of selected

Philippine industry sectors influences the over-all stock market price

behavior as measured by the Phisix. Some industries are necessarily more

determinate of overall stock price movement than other industries.

It is argued that the risk behavior of the Philippine Stock Market

Index is significantly influenced by the behavior of the index levels of the

industry sectors composing the Philippine Stock Market. The expected

signs of the relationships of the indices to the Phisix are all positive. It is

expected that all sectors contribute significantly and positively to the

general movement of the Philippine stock market.

The research sought to reject the null hypothesis that there is no

relationship between volatilities in the price of the respective industry sector

to the volatilities in the Philippine Stock Exchange composite index.

Rejecting the null hypothesis allowed the alternative hypothesis to be

 13

accepted, which is that volatilities in the industry indices explain the

volatilities in the Phisix.

1.5 Assumptions

This thesis considers many of the defining and exploratory theories

that drive research and studies in the areas of risk and finance. Because

multiple regression is utilized to analyze raw data and establish the risk

behaviors of industry stocks, the obvious assumption in this case is the

linearity of relationships between the variables studied.

This thesis did take off from the basic CAPM theory which assumes

linearity of relationships between variables.

There is an expectation that the study may reveal results contrary to

what is expected. However, the expected signs in the beta coefficients of

the estimation model may take a positive or negative value. Thus, the

estimation model are structured with all positive signs in the linear

regression form. It is assumed that the stock indices take on a positive

relationship with the Phisix as the dependent variable in the regression

analysis.
 14

Since the variables are stock price indices, the objection to the linear

model that makes the CAPM a good starting point for explaining stock price

and risk behaviors should be dealt with.

There should be a linear relationship between the variables that are

considered in this study. The non-linear exogenous effects of other

variables enter the CAPM-derived Philippine model via the inclusion of the

error term for typical regression equations, as shown in Harvey (1995).

Risk is viewed in this study as synonymous to the concept of the

beta coefficient in regression analysis. The CAPM theory has demonstrated

that the beta coefficient is a measure of risk. It is the standard in stock

market analysis to view the beta coefficient as a measure of risk although

estimated from an entirely different set of data variables in the case of firm-

level riskiness.

1.6 Significance of the Study

The thesis looked into the relationships of industry equity indices to

the Phisix. The significance cannot be overemphasized. There is a need to

identify the risk contributions of specific industry sectors to the volatility in

the Philippine stock market as well as to validate the strength of the Phisix

as the true measure of the health of the Philippine stock market itself. This
 15

will have interesting implications for investment decisions and financial

market analysis with perhaps policy ramifications as well.

For the most part, from a purely academic viewpoint, the study

attempts to fuse together in a simple business-oriented approach the

theories that establish the beta coefficient as the principal risk criterion as

far as equity markets analysis is concerned. The study validated its

analytical tool as one measure for assessing the strength of Philippine

business and industry as it contributes to financial capital generation and

consolidation or capital formation. The beta coefficient was considered for

each industry sector. The interplay of beta coefficients in this exercise

gauged the extent with which the respective sector exerts influence on the

Philippine stock market. The impact of the political environment as well as

macroeconomic phenomena such as the Asian Financial Crisis which hit

the Philippines during the time considered in this study provided insights as

well.

1.7 Scope and Limitation

The use of beta coefficient as a standard measure of market risk is

probably overemphasized. Some studies even done locally have pointed

out the limitations of the CAPM model, which rests on assumptions of

 16

linearity in explaining stock pricing behavior which may even exhibit non-

linear relationships.

In this study, the relevant measure of market risk as denoted by the

industry beta is determined by establishing the relationship between stock

price movements (as measured by the stock indices) rather than the

CAPM-validated returns on a security or a portfolio of securities and on the

market portfolio.

The study is more of evaluative rather than exploratory, although it

does attempt to test the validity of the Phisix as representative of the stock

market in general when only the 30 best-performing stocks are represented

in that basket of stocks.

It is the intent of the proponent to provide insightful analysis on the

contributions of industry sectors to stock market development and in

considering the risk exposure when investing in certain industry stocks.

The study did not primarily attempt to advocate or debunk a theory. It

does not seek to argue that either accounting fundamentals or economic

fundamentals or external shocks contribute to the volatilities in the industry

stock prices or the stock market itself.

What it attempted to look into is to consider the measure of the

economic beta coefficient to capture the risk behavior of portfolio indices

 17

within the Capital Asset Pricing Model theory framework. This means

assessing purely market risk influence contributed by specific industry

sectors active in the Philippine Stock Exchange.

The position that the proponent takes in all these debates on the

efficacy of the beta criterion in measuring non-linear relationships exhibited

by stock markets is that the beta criterion is proposed to estimate the

determining relationships between the variables at hand. It is argued that if

the linear estimation is seen as valid and the beta criterion establishes the

relationships between variables, then the riskiness of the Philippine stock

market to industry sector performance, as measured by the respective beta

coefficient should be valid as well.

The argument of autocorrelation for this regression equation did not

hold because the stock market index, the Phisix, is determined in actuality

by the values of the market capitalization reflecting price volatility of only

thirty (30) carefully-selected stocks of Philippine companies.

The study considered purely aggregate data of industry sectors. No

attempt was made to look into firm-specific data or relationships. Thus, the

estimation model may not be the appropriate model with which to derive

firm-specific conclusions.
 18

1.8 Definition of Terms

All Shares Index. A complementary index to the Phisix (or Phisix) which

takes into consideration stocks of all listed companies except those

suspended, as well as bonds, preferred and warrants.

Beta Coefficient. A measure of the responsiveness of the expected return

on a particular financial security relative to movements in the average

expected return on all other securities in the market. The Financial Times

All-Share Index or the Dow-Jones Industrial Average Index are usually

taken as proxy measurements for general market movements.

In the capital-asset pricing model, the beta coefficient is taken as a

measure of the market (or non-diversifiable) risk of a particular security. The

beta coefficient links the return on the security and the average market

return.

Capital asset pricing model. A model developed to determine the required

rate of return for an investment that considers the fact that some of the total

risk associated with the investment can be diversified away; in essence, the

model suggests that the risk premium associated with an investment should

only be based on the risk that cannot be diversified away rather than the
 19

total risk; investors should not be rewarded for not diversifying-that is, they

should not be paid for taking on risk that can be eliminated through

diversification.

Capital formation. Capital is said to be "formed" when savings are used for

investment purposes often investment in production. It basically refers to

the net additions to the (physical) capital stock in an accounting period, or,

to the value of the increase of the capital stock; though it may occasionally

also refer to the total stock of capital formed. It refers to the creation of

capital. For example, capital is created when banks lend the money they

hold in savings accounts to firms that use the money to purchase

machinery.

Commodity risk. A type of market risk or risk that commodity (grains,

metals) prices will change.

Currency risk. A type of market risk or risk that foreign exchange rates will

change.

Equity risk. A type of market risk or the risk that stock prices will change.
 20

Equity index risk. A type of market risk or the risk that a stock index will

change.

Interest rate risk. A type of market risk or the risk that interest rates will

change.

Market risk. It is the risk that the value of an investment will decrease due

to moves in market factors. This risk is also known as systematic risk or

relevant or non-diversifiable risk.

Political environment. In this research, political environment denotes the

political system, structure, stability, ideology espoused under the three

government administrations of the Philippines during the period in the

study, namely, the political administrations of Fidel V. Ramos ( January

1997-May 30, 1998), Joseph E. Estrada (June 30, 1998-January 20, 2001 )

and Gloria Macapagal-Arroyo (January 20,2001-present ).

Political risk. The risk of loss when investing in a given country caused by

changes in a country's political structure or policies, such as tax laws,

tariffs, expropriation of assets, or restriction in repatriation of profits. It is a

 21

broad term to collectively describe the risks companies and investors face

due to the exercise of political power.

Portfolio Risk. By combining more and more investments that are not

perfectly correlated-that is, do not mirror each others’ movements on a

relative basis-to form a portfolio, the risk of the portfolio can be reduced; the

amount of the risk reduction depends on how the investments in a portfolio

are related-risk can be eliminated if two investments that are perfectly

negatively related are combined to form a portfolio; the smaller the

relationship among the various investments included in a portfolio, the

greater the reduction of risk, or diversification; to manage, thus reduce, risk

investors should diversify.

Risk. It is a concept that denotes a potential negative impact to an asset or

some characteristic of value that may arise from some present process or

future event. In finance, risk is the probability that an investment's actual

return will be different than expected.

Riskiness. This term is analogous to the beta coefficient, which measures

the riskiness or the characteristic of being risky of a variable, in this case,

the Phisix, to movements in the market, in this case, the sectoral indices.
 22

The Philippine Stock Exchange, Inc. (PSE or Exchange) is a private

organization that provides and ensures a fair, efficient, transparent, and

orderly market for the buying and selling of securities.

The PSE Composite Index (Phisix, PSEi). An aggregate measure of

relative changes in the market capitalization of common stocks that

provides comprehensive picture of market trends. It is composed of a fixed

basket of 30 listed common stocks carefully selected to represent general

movement of market prices. The Phisix, renamed the PSEi in 2006, is a

market-value-weighted index of the shares of 30 representative companies

from different sectors of the local bourse. Each industry is represented by a

company with the highest market capitalization within the industry.

Volatility. Refers to the degree of (typically short-term) unpredictable

change over time of a certain variable as in price volatility.

 23

Chapter II

Review of Related Literature

2.1 Review

The concepts embraced in this theory rest on the Capital Asset

Pricing Model (CAPM) independently developed by Jack Treynor in 1961,

William Sharpe in 1964, John Lintner in 1965, and Jan Mossin in 1966, as

offshoot to Harry Markowitz’s work in 1952 on diversification and modern

portfolio theory. The beta coefficient as a measure of market risk is widely

associated with the development of the CAPM.

Diacogiannis and Makri (2008) state that the relevant measure of risk

is beta, which is the slope parameter of a linear regression equation relating

the security’s returns to the corresponding market returns. They said that

beta has received considerable attention by being widely applied in several

topics including the measurement of the cost of equity for capital budgeting

and firm valuation and evaluation of security or portfolio performance. The

method of ordinary least squares (OLS) is typically employed to estimate

security beta. However, a number of beta estimates can emerge for a

security depending on various considerations such as the choice of the

market index, the length of the return measurement interval, and the

sample period.
 24

In Brigham and Young’s (2004) textbook illustration, the CAPM

relationship is depicted as follows:

ki = krf + (km-krf)βi;
where ki is the required return on stock i,
krf is the risk-free rate,
(km-krf) is the market risk premium, and
βi is the measure of risk in relation
to the returns on a security or portfolio
of securities to the returns on the market
(stock i’s β).

This is very close to the simple model presented by Bradfield (2003)

when he proposed that some stocks are more sensitive to movements of

the overall market index, and the beta coefficient represents this measure

of riskiness of some stocks to changes in the overall market index.

Vides (2007) posits that the CAPM implies a positive linear

correlation between expected return and beta of the security and stocks

with larger beta will demand higher expected return than stocks with smaller

beta.

Bradfield (2003) proposes that the concept of beta arises because all

stocks tend to move to some extent with movements in the overall market.
 25

Some stocks tend to move more than others when the market moves; thus,

their sensitivity to the movements of the overall market index is an

important measure, known as the beta coefficient. He expounded that the

beta coefficient is estimated by running a market model regression, which

is expounded further in this study in a separate chapter.

Chari and Henry (2004) recall Sharpe’s 1964 observation that the

levels of expected stock returns should vary cross-sectionally according to

the degree of firm exposure to systematic risk. However, they pointed out

that research from the last several years provide little empirical evidence to

suggest this prediction. Both authors jointly attempted to study whether risk

matters in asset pricing, this time by examining stock market liberalization

behavior, in selected countries in Asia and Latin America. They revealed

that liberalization in these countries reduces systematic risk as it changes

the relevant source of systematic risk from the local stock market index to

the world stock market index.

Michailidis, Tsopoglou, Papanastasiou and Mariola (2006) state that

the CAPM suggests that high expected returns are associated with high

levels of risk. They articulated that the CAPM argues that the expected

return on an asset above the risk-free rate is linearly-related to the non-

 26

diversifiable risk measured by the asset’s beta. They added that although

the CAPM is the basis of modern portfolio theory, accumulating research

has cast doubts on the model’s ability to explain the actual movement of

asset return. The authors discuss the empirical evidence on the validity of

the CAPM as tested previously by Black, Jensen, and Scholes in 1975, and

by Fama and McBeth in 1973.

Fama and French (2006) cite the inadequacies of the CAPM in

predicting asset returns using market betas for variations in β and the book

value-to-market value ratio that are unrelated to size. They said that the

CAPM model failed to predict asset returns for the given periods.

Michailidis et al. (2006) note that the CAPM theory has been

challenged empirically as well. In their paper, the authors presented the

findings of Banz in 1981 where it was shown that firm size explain cross-

sectional variations in average return on a particular collection of assets

than beta. They said that Fama and French in 1979 found no relation

between risk and return even though the latter utilized the same procedure

as the study earlier conducted by Fama and McBeth.

 27

Park (2004) maintains that while most CAPM tests following the

Pettengill procedure focused on the cross-sectional data, it is more

appropriate to examine the conditional relation between beta and return by

using time-series analysis since beta is not stable over time. He

summarized the CAPM theory by reiterating that the expected return on any

risky security or portfolio of risky securities can be measured by the risk-

free rate and the market risk premium multiplied by the beta coefficient.

Odabasi (2000) maintains that beta stability increases as the portfolio size

becomes larger.

Campbell, Polk and Vuolteenaho (2005) cite that in the CAPM, the

risk of each stock is measured by its beta with the market portfolio and it is

natural to ask whether betas are determined by shocks to cash flows or

discount rates. They note that common variation in stock prices is

particularly important when it affects the measures of systematic risk that

rational investors use to evaluate stocks. They propose that if stocks are

priced by discounting their cash flows at a rate which is constant over time,

although possibly varying across stocks, then movements in stock prices

are driven by news about cash flows. In this case, common variation in

prices must be attributable to common variation in cash flows. If discount

rates vary over time, however, then groups of stocks can move together

because of common shocks to discount rates rather than fundamentals.

 28

Campbell et al. (2005) claim that in the extreme, irrational investor

sentiment can cause common variation in stock prices that is entirely

unrelated to the characteristics of cash flows. They cite the works of

Barberis, Shleifer and Wurgler (2005) and Greenwood (2005) which

suggested that this explains the common movement of stocks that are

included in the S&P 500 and Nikkei indexes.

Locally, Bautista (2003) earlier demonstrated in his study on stock

market volatilities in developing countries like the Philippines that stock

markets are sensitive not only to economic activity but also to changes in

the political and international economic environment. Bautista (2005) noted

in his study of stock price data of seven economies that included Hong

Kong, Indonesia, Thailand and the Philippines that global and regional

events like the 1990 Gulf War and the 1997 Asian Currency Crisis led to

high volatility episodes whose magnitude relative to normal times differ from

country to country. In addition, he said that country-specific events like the

opening up of country borders also led to high volatility periods.

In recent years though, there has been a low level of price volatility

over a wide range of financial assets and markets (BIS Study Group, 2006).

The BIS Study Group (2006) reported that from mid-2004 to March 2006,
 29

there has been low volatility of short-term and long-term interest rates,

stocks, exchange rates, and corporate spreads as compared to the

previous 5 to 10 years in both industrial countries and emerging market

economies. Volatility was noted to be low for a prolonged period

simultaneously across different assets and markets in both industrial

economies and emerging market economies. The firm-specific components

of volatility seem to have become more important over time. The theoretical

and empirical results surveyed in this report suggested that volatility is

negatively related to firm profitability and positively related to leverage and

to uncertainty about profitability.

De Ocampo (2004) establishes the role of the beta coefficient in

explaining returns in the Philippine equity market arguing that a given stock

portfolio may perform better when invested in high beta stocks when the

market is up and in low beta stocks when the market is down.

Campbell et al. (2005) indicate that cash flow fundamentals of

growth and value companies determine high betas of growth stocks and

value stocks. They submit that accounting measures of firm-level risk have

predictive power for firms’ betas with market-wide cash flows and this

predictive power arises from the behavior of firms’ cash flows. The
 30

systematic risks of stocks with similar accounting characteristics are

primarily driven by the systematic risks of their fundamentals.

2.2 Synthesis

For the most part, the literature in the area of risk analysis of the

financial markets, especially relating to risk and returns in stock market, has

a long history. There has been much discussion on the merits and

limitations of the Capital Asset Pricing Model, specifically the beta

coefficient as a measure of risk, in explaining behavior of stock market

returns and stock market volatilities.

The defining works on the Capital Asset Pricing Model and the

elaboration on these works by later authors provide very useful and critical

tools to the analysis of how specific Philippine industry sectors behave and

influence the general direction and even stability of the Philippine stock

market as represented by thirty equity stocks. The proponent deviates from

documented theories and studies by arguing a case for the riskiness of the

market index on one hand to the general movement this time of the industry

sectors that make up the Philippine stock market.

 31

This contrasts with the popular CAPM theory that the market risk of

an individual security is the appropriate measure of risk for that security

since non-market related risk can be eliminated by the aggregation of

securities in a portfolio.

In this instance, the relevant measure of risk is the market risk of the

Philippine stock market to the volatilities in the stock price movements of

the industry sectors that compose it.

In general, it is argued that the beta model as a measure of risk

should hold for the Philippine stock market and that the relationships

between the variables (industry indices versus the Phisix) are linear in

nature.
 32

Chapter III

Conceptual/Operational/Theoretical Framework

Based on the appreciation of the literature on measuring market risk,

this research focuses on analyzing the risk behavior of the Philippine stock

market with respect to the behavior of the industry indices.

Put in a more mathematical context, the study looks into the sensitivity

of the Philippine stock market (measured by the Phisix representing the 30

best-performing Philippine stocks) to movements in the industry sector

stocks. This establishes the relative strengths of particular industries with

respect to explaining the robustness, if at all, of the stock market.

The beta coefficients derived for the linear regression developed for

this study would measure the risk relationship of the Philippine stock market

to the particular industry sector.

The idea is that such identification of the risk relationships of industry

stocks to the Phisix will bring to the fore the extent of risk exposure of the

Phisix to volatilities in industry stock prices and, therefore, to volatilities in the

business management of firms of their respective market risks.

 33

The conceptual and operational framework is illustrated here:

Figure 2. Conceptual and Operational Framework

Independent Dependent
Variables Variable

Financial Index
Industrial Index PSE Composite
Property Index Index
Mining Index (Phisix)
Oil Index

Intervening Variables

Political environment
Asian financial crisis

Theoretically, the regression model tested through Generalized Least

Squares (GLS) estimation is as follows:

Phisix = α + β1Industriali + β2Financiali + β3Propertyi +

β4Miningi + β5Oili + β6Dummyi +ε,

where ε measures the firm-specific or non-diversifiable risks, and

Dummy is 1 if the period is under GMA administration
 34

The beta coefficients generated in the testing of the regression model

represent the measure of risk, chiefly, market risk. This model measured the

extent to which the selected industry indices moved together with the Phisix.

As of 1 May 2007, the Philippine Stock Exchange has a total of 241

listed companies. Stocks listed in the PSE are classified into six sectors;

namely, Financials, Industrial, Holding Firms, Property, Services, and Mining

& Oil. Companies are classified according to the business that generates the

bulk of their revenues.

Table 1.
Industry Sectors in the Philippine Stock Exchange
(As of 1 May 2007)

No Industry Business Activity

Sector
1 Financial Banking; Investments; Finance
Electricity; Energy; Power & Water; Food, Beverage
2 Industrial & Tobacco; Construction; Infrastructure & Allied
Services; Chemicals; Diversified Industrials
Holding
3 Diversified companies in 3 or more industries
Firms
4 Property Land and property development
Media; Telecommunications; Information
5 Services Technology; Transportation services; Hotel &
Leisure; Education; Diversified services
Mineral extraction; Oil exploration, extraction and
6 Mining & Oil
production

Source: www.pse.com.ph
 35

In analyzing the risk behavior of the Phisix against key sectors, the

proponent was limited to the consideration of sectoral indices for which

suitable data are available. As such, the defining sectors followed the

sectoral grouping that was formerly adopted and for which a 9-year data

were gathered by the Philippine Stock Exchange. These sectors are as

follows:

Table 2.
Industry Sectors in the Philippine Stock Exchange
(As of 2004)

No Industry Business Activity

Sector
1 Financial Banking; Investments; Finance
Holding Firms; Telecommunications; Power and
Energy; Food, Beverage & Tobacco; Construction;
2 Industrial
Transportation Services; Manufacturing; Distribution
and Trading; Hotel, Recreation and Other Services
3 Property Land and property development

4 Mining Mineral extraction

5 Oil Oil exploration, extraction and production

Source: The Philippine Stock Exchange, Inc. PSE Factbook 2004.

 36

Chapter IV

Methodology

This paper utilized multiple regression analysis using generalized least

squares estimation in testing the hypothesis that the daily trading activities of

the Philippine stock market, as measured by the Phisix, is influenced by the

behavior of the industry sector stock price indices. Therefore, the

performance of companies in specific industry sectors, as far as their

management of their respective market risks are concerned, contribute to the

volatilities in the Phisix.

There is undoubtedly an assumption of serial fluctuations in

measuring the relationship of the daily trading behavior of the Philippine

stock market to the movements of selected industry sector indices. The

degree to which a particular industry stock index competes in the market

evidently pushed up or pulled down the Philippine stock market.

4.1 Research Design

The researcher used the descriptive research design. This method

describes the nature of the situation as it exists at the time of the study and

explores the causes of particular phenomena. The process involves

describing, analyzing, and interpreting the data that now exists. Thus, it
 37

shows comparison and contrast, and attempts to discover relationship

between existing non-manipulated variables. Under the descriptive research

design, correlation and time series analysis were used to explain the

Philippine Stock Exchange phenomena in the multiple regression process.

4.2 Setting

The study looked into the influence of five selected industry sectors

listed in the Philippine stock market; namely, the financial sector composed

mainly of banks, industrial sector composed mainly manufacturing firms,

property sector, mining, and oil companies. Complete data for all relevant

indices are available for the period 1997-2005.

4.3 Participants

Data which capture the industry stock price index were collected from

the Philippine Stock Exchange. Five sectors were considered in this study:

(1) financial sector, (2) industrial sector, (3) property sector, (4) mining

sector, and (5) oil sector. The unavailability of indicative stock price indices

for the services sector and holding firms for the required period of study

preclude their inclusion in the model that has been developed. Table 3 shows

the publicly-listed companies by sector.

The Philippine Stock Exchange Composite Index, the PSEi or Phisix

as it was known during this period represents the thirty best-performing

Philippine stocks. These 30 best performers are presented in Tables 4 & 5.

 38

The sectoral indices were analyzed against the PSE composite index (Phisix)

composed of a basket of thirty (30) selected securities. Time series analyses

were conducted to determine the daily performance of the Philippine stock

market against the above mentioned 5 sectors.

Table 3. Sectoral Index Composition

(1998, partial)
No. Sector Publicly-Listed Companies
1 Financial BPI Union Bank
PCIBank Security Bank
(10 Metrobank Philippine Savings Bank
Stocks) Philippine National Bank Bankard, Inc.
Far East Bank & Trust Co. Vantage Equities, Inc.
2 Industrial San Miguel Corporation Pilipino Telephone Corp.
Ayala Corporation Benpres Holdings Corp.
(50 First Phil. Holdings Petron Corporation
Stocks) Meralco ABS-CBN Broadcasting
PLDT Aboitiz Equity Ventures
Filinvest Development Corp. Universal Robina Corp.
JG Summit Holdings, Inc. Jollibee Foods Corp.
3 Property Ayala Land, Inc. Filinvest Land, Inc.
SM Prime Holdings Empire East, Inc.
(10 C&P Homes, Inc. Primetown Property
Stocks) Megaworld Corporation Universal Rightfield
Fil-Estate Land, Inc. Uniwide Holdings, Inc.
4 Mining Philex Mining Corp. United Paragon Mining
(7 Lepanto Consolidated Mining Atlas Consolidated
Stocks) Abra Mining Itogon-Suyoc Mines, Inc.
Manila Mining
5 Oil The Philodrill Corp. Trans-Asia Oil & Mineral
(6 South China Petroleum Alcorn Petroleum
Stocks) Oriental Petroleum Vulcan Industrial

Source: Philippine Stock Exchange, Memo for Brokers, February 1998

 39

Table 4. Companies Comprising the Philippine Stock Exchange

Composite Index
(As of 16 November 2007)

Company Stocks Comprising the Philippine Stock Exchange

Composite Index
1 Aboitiz Equity Ventures, Inc.
2 ABS-CBN Broadcasting Corporation
3 Ayala Corporation
4 Ayala Land, Inc.
5 Banco De Oro-EPCI, Inc.
6 Bank of the Philippine Islands
7 Belle Corporation
8 DMCI Holdings, Inc.
9 Filinvest Land, Inc.
10 First Gen Corporation
11 First Philippine Holdings Corporation
12 Globe Telecom, Inc.
13 Holcim Philippines, Inc.
14 International Container Terminal Services, Inc.
15 JG Summit Holdings, Inc.
16 Jollibee Foods Corporation
17 Lepanto Consolidated Mining Company
18 Manila Electric Company
19 Manila Water Company, Inc.
20 Megaworld Corporation
21 Metropolitan Bank & Trust Company
22 Petron Corporation
23 Philex Mining Corporation
24 Philippine Long Distance Telephone Company
25 PNOC-Energy Development Corporation
26 Robinsons Land Corporation
27 San Miguel Corporation
28 SM Investments Corporation
29 SM Prime Holdings, Inc.
30 Universal Robina Corporation

Source: The Philippine Stock Exchange, Memo for Brokers. October 2007.
 40

Table 5. Companies Comprising the Philippine Stock Exchange

Composite Index
(As of 1 December 2005)

Company Stocks Comprising the Philippine Stock Exchange

Composite Index
1 Aboitiz Equity Ventures, Inc
2 ABS-CBN Broadcasting Corporation
3 Ayala Corporation
4 Ayala Land, Inc.
5 Banco de Oro Universal Bank
6 Bank of the Philippine Islands
7 Belle Corporation
8 Benpres Holdings Corporation
9 Digital Telecommunications Phils., Inc.
10 DMCI Holdings, Inc.
11 Equitable PCI Bank, Inc.
12 Filinvest Land, Inc.
13 First Philippine Holdings Corporation
14 Globe Telecom, Inc.
15 Holcim Philippines, Inc.
16 International Container Terminal Services, Inc.
17 Jollibee Foods Corporation
18 Lepanto Consolidated Mining Company
19 Manila Electric Company
20 Manila Water Company, Inc.
21 Megaworld Corporation
22 Metro Pacific Corporation
23 Metropolitan Bank & Trust Company
24 Petron Corporation
25 Philex Mining Corporation
26 Philippine Long Distance Telephone Company
27 Pilipino Telephone Corporation
29 San Miguel Corporation
28 SM Investments Corporation
30 SM Prime Holdings, Inc.

Source: The Philippine Stock Exchange, Memo for Brokers 274-2005.

 41

4.4 Research Instrument

The testing of hypothesis involved secondary data analysis of actual

price movements as measured by the Phisix as well as particular sector

indices. These data are available from the Philippine Stock Exchange. Data

observations on a daily basis within a nine - year period from 1997-2005

were collected for the purpose of running the regression model.

4.5 Procedure for Data Gathering

The study primarily relied on secondary data gathered from daily stock

price indices (industry and Phisix) that are readily available in the Philippine

Stock Exchange at the time of the study. A purposive sampling was used to

select the five (5) key industries that are players in the Philippine stock

market. The criteria for selecting the key industries are as follows: (1) the

companies should belong to the top performers in the stock market; (2) they

must have a complete set of daily data from January 1997 to December

2005; and (3) there must be no negative values in their daily stock index

within the test period. It is argued that the relative strengths of these

industries contribute to the robustness or weakness, as the case may be, of

the Philippine stock market. A total of 2349 days were collected as samples.
 42

4.6 Derivation of Raw Data

The Philippine Stock Exchange Composite Index and the sector

indices were calculated using the weighted market capitalization method

which measures the relative changes in market capitalization of common

stocks. In the case of the Phisix, full market capitalization was considered in

the calculation of the PSE Composite Index. In 2006, the PSE shifted to the

free float market capitalization to account for only those shares that are

available for trading and investing by the public.

The index is calculated by dividing the total current market

capitalization of constituent stocks (the sum of the free-float or outstanding

shares and the current market price of each component stock) with the total

base market capitalization (computed using the base market price).

The general formula for the indices is:

PSE Indices = Total Current Market Capitalization x 100

Total Base Market Capitalization

The PSE adopted a chain-linking method to prevent data series

disruption resulting from adjustments such as stock conversions,

reclassification of companies among sectors, addition and deletion of

constituent stocks to the indices, and corporate restructuring. The index in

 43

effect consists of a number of chain-linked series where each link was

established at the point in time the changes occurred. The daily chaining

method is:

Today’s Index = Yesterday’s Index x Today’s Market Capitalization ,

Yesterday’s Market Capitalization

where, Market Capitalization = Closing Price * Free-Float Shares

The criteria for the selection of stocks for the sector indices are the

same as that in the Composite Index, only the parameters are more relaxed.

Only common shares of domestic companies listed for at least six (6) months

are qualified to be included in the basket of index stocks.

For the sector index, constituent stocks must have at least P1 million

average daily turnover and should be tradable for at least 75% of the total

trading days. Only big companies ranked according to (free float) market

capitalization are considered in the index.

Prior to the shift to free float index, the PSE Composite Index

composition criteria was based on liquidity (average at least P5 million daily

value turnover), tradability (tradable for at least 95% of the total trading

days), and only big companies ranked according to full market capitalization

are considered for inclusion in the index.

 44

4.7 Model Specification:

Based on the objectives and the conceptual framework of the study,

the general form of the Philippine Stock Exchange Index (Phisix) model of

the study is expressed as:

Phisix = a + b1 (F) + b2(I) + b3 (M) + b4(O) + b5(P) + b6(D1) (1)

where:

Phisix = Philippine Stocks Exchange Index

F = Finance sector (index)

M = Mining Sector (index)

I = Industry sector (index)

O= Oil sector (index)

P= Property sector (index)

D1 = is a dummy variable representing the political

influence of the GMA administration within January

21, 2001 – December 2005). That is,

= 1, if the period covered the administration of GMA

0, if the period were those of others’ administration

a, b1, b2, b3, b4, b5 and b6 = parameters to be tested for significance.

 45

The above Phisix model depicts numerical or quantitative and

qualitative variables. The quantitative variable (finance, industrial, mining, oil

and property) numerically influenced the daily activities of the Philippine

Stock market (Phisix). However, the variation in the Philippine Stock

Exchange index reflects different characteristics of the environment that can

affect their daily performance. These fluctuations or variations indicate the

presence or absence of qualities or attributes that affect daily trading

activities. This qualitative variable is the dummy variable (D1) that represents

the influence of characteristics in the daily Phisix.

Maddala (2001) explains some of the usefulness of dummy variables

are explaining differences of intercepts terms and testing stability of

regression coefficients. Maddala showed that a single dummy variable

(month of December) influences the rest of the periods. He suggested the

use of a single variable if a researcher feels that the particular dummy

variable makes a big difference in the model. In this study, the researcher

feels that the number of Phisix trading days under a political administration

(January 1997 – December, 2005) strongly affects the push up in Phisix

activities.

It must be noted that the business sectors were critical to the morality

issue and management capability of President Joseph E. Estrada (June 30,

1998 – January 20, 2001). The coverage of President Fidel V. Ramos’

 46

administration (January 1997 – May 30, 1998) was very short in the test

period. Hence, in the regression, the above mentioned administrations

served as the benchmark (dummy = 0, other administrations) in the

regression.

The dummy variable D1 takes the values 1 or 0 --- 1 representing the

period of a political administration (GMA) from January 21,2001 to December

31, 2005 and 0 representing the other political administration(s). As a rule in

using dummy variable, there should be (m – 1) dummy variable in the model

to avoid the situation of perfect multicollinearity (Gujarati, 1999). This means

that only one (1) parameter should be introduced in the model (2 – 1 = 1) to

avoid the dummy variable trap.

If political administration affects daily activities of Phisix then equation

(1) can’t be used straightforwardly for prediction because its structure is not

stable (Gujarati, 1999). That means, within the equation, the daily trading

Phisix varies (as influenced by the economic system) in every change of

political administration. Therefore, the Phisix prediction equation per

influence of political administration can be expressed as:

Average Phisix under the term of GMA ( D1 = 1):

A[Phisix | GMA = 1, Xi] = (a + b6) + biXi (2)

 47

Average Phisix not under the term of GMA (D1 = 0):

A[Phisix | GMA = 0, Xi] = a + biXi (3)

Where:

A = average

Phisix = Philippine Stock Exchange index

Xi = any sectors (financial, industrial, mining, oil, and

property) that affects Phisix;

i = 1, 2, 3, 4, 5, 6.

GMA = dummy variable representing GMA’s administration

Equations 2 and 3 are nonstochastic (values are fixed) and assumes

that the error term (ui) is equal to 0; A(ui = 0).

4.8 Statistical Tests

1. Test for the individual parameters

The parameters (a, b1, …, b6) are tested for significance at 5% - 10%

level. The parameter test determines the order of factors’ effect to the daily

trading activities of Phisix. value of the t-distribution.

The computed t-value is significant if its value is greater than the

critical t-value. This means the hypothesis (theory) that the trading sectors
 48

(finance, industrial, mining, oil and property) have no effect to the Phisix (bi =

0) is not true and the research hypothesis (bi ≠ 0) is accepted. Since the

research is not expecting a theoretical sign for each respective factor, the t-

test conducted is two-sided or two-tailed (Gujarati, 1999).

2. Test for the goodness-of-fit

The whole regression model can be statistically tested for its overall

significance through the analysis of variance (ANOVA). This is done by

determining the source of variation (due to regression and due to residual or

error). This means that the first source explains the variance explained by the

five (5) sectors and the second source considers the unexplained variance

(residual or error). These sources of variations are influenced by their

respective degrees of freedom (d.f.). The explained variation (due to

regression) has (K – 1) degrees of freedom in the numerator and the

unexplained variation (due to error) has (n – K) degrees of freedom in the

denominator. The F-ratio is the result of dividing the two sources of variation.

The F-ratio measures the overall significance of the Phisix regression

model and it also tests the significance of R2. If it is equal to zero, (R2 = 0),

then the F-ratio is also 0. Thus, the overall significance of the whole

regression model indicates a good prediction model for the Phisix.

 49

Appropriate model for predicting the effect of industry stock price

indices on the Phisix is based on the following reference criteria of Harvey

(1981) as cited by Gujarati (1999):

a. Parsimony = a model must be kept as simple as possible.

b. Identifiability = only one estimate for every parameter.

c. Goodness of fit = adjusted R-square is as high as possible.

d. Theoretical consistency = correct signs for the regression

coefficients.

e. Predictive power = the predicted values (theoretical predictions)

should be comparably close to the actual values.

This is to say that the study seeks to reject the hypothesis that there is

no relationship between the dependent variable as represented by the

Phisix and the respective independent variables as represented by the

sectoral indices.

4.9 Data Analysis

Secondary data analysis of 2349 samples (days of trading) from

sectors and PSE were utilized for the multiple regression model using

statistical software packages such as Statistical Package for the Social

Sciences (SPSS).
 50

Chapter V

Results and Discussion

This section presents the observation and discusses the results of the

statistical analyses. The results include statistically significant business

industry sectors that influence the fluctuations in the Philippine stock market

(Phisix). The performance of these selected thirty (30) best-performing

Philippine stocks in five specific industry sectors makes the Phisix stronger

(or weaker) everyday. The combined performance of the selected companies

simplified the interplay of capital investment in the Philippines.

The research focuses on the problem of measuring the influence of

Filipino business and industry on the growth of the Philippine stock market as

they manage market risk (implicitly) in their respective areas of business

expertise.

1. Determine the statistically significant selected industry or

industries that influence the movement of the Philippine stock

market (Phisix), the companies making up these sectors and the

effect of these industry indices on the Philippine stock market

indicator.
 51

The selected industries were composed of financial, industrial,

property, mining, and oil sectors. In 1998, the financial sector index was

represented by stocks of companies in the banking and financial services

industry, namely: BPI, Metrobank, PCI Bank, Bankard, Inc., Security Bank,

Union Bank, Philippine National Bank, Far East Bank & Trust Co., Vantage

Equities, Inc., and Philippine Savings Bank. These notable banks provided

capital investment in the form of service and loan transactions that were

absorbed in the other industries for the enhancement of productive

endeavour.

The industrial sector is known for its products like beverage and

spirits, food, energy, steel, and others. In 1998, this sector’s index included

the following company stocks: San Miguel Corporation, Ayala Corporation,

First Philippine Holdings, Meralco, Universal Robina Corporation, Jollibee

Foods Corporation, Manila Water Co, Inc., Holcim Phils, Inc., Petron

Corporation, EEI Corporation, Asian Terminals, Inc. and Southeast Asia

Cement Holdings. Most products of the companies were consumed in the

property sector (including households and private sector).

The property sector represents the flow of stocks from its commercial

and residential activities to the other sectors of the economy. These

companies provided retail and wholesale trading of products strategically

located in a demand driven world of prime land (supermalls). Residential

 52

homes and commercial offices were included in superbly-designed

architectures of properties suited for middle and high income families.

Enumerated under this sector as constituent stocks of the property index

were the following: Ayala Land, Inc., SM Prime Holdings, C&P Homes, Inc.,

Megaworld Corporation, Primetown Property Group, Filinvest Land, Inc.,

Empire East Inc., Uniwide Holdings, and Universal Rightfield Property

Holdings.

The mining sector harnessed the non-renewable resources of the

economy. Extraction of these non-renewable resources (i.e. copper, gold,

nickel, iron and others) needed capital stocks to sustain operations.

However, a critical trade-off for environmental conservation and economic

use of resources ensues in the extraction process. Thus, the investment in

this sector is highly risky but is rewarded with high returns. High risks usually

compelled investors to look for other less risky investment portfolio. In 1998,

this sector’s index included the stocks of the following companies: Philex

Mining Corporation, Lepanto Mining, Manila Mining, Atlas Consolidated, Abra

Mining, United Paragon, and Itogon-Suyoc Mines.

The oil industry supplied the energy requirements of the economy in

the form of natural gas, crude oil, refined products (gasoline, petrol,

lubricants, octane and others), and by-products (asphalts, bunker oil, and

others). Majority of the supply of oil came from abroad that increases its price
 53

every period. Through partnership with foreign companies, the Philippines

explored its northern territory for oil to be self-reliant in energy. This deep-sea

exploration needed capital stocks to sustain its activities.

However, the volume of oil extracted in the exploration was not enough to

supply the increasing local demand for oil and oil products. Also, the quality

of the locally – extracted oil is not of exportable grade. For these reasons, the

investment for oil exploration is highly risky and not bullish. The following

companies were involved in oil exploration at the time covered in this study

and their stocks were the constituent stocks under the oil index in 1998,

subject to review by the PSE: The Philodrill Corporation, Oriental Petroleum,

South China Petroleum & Exploration, Trans-Asia Oil & Mineral Development

Corporation, Alcorn Petroleum & Minerals Corporation and Vulcan Industrial

& Mining Corporation.

A generalized least-squares regression equation was used to

determine the statistically significant sectors that influence the behavior of

the Philippine Stock Exchange Index (Phisix). Under the autoregressive –

time series process, the Prais-Winsten method was used to comply with the

statistical criteria in selecting a good model (SPSS Help Option, 1998). The

regression results showed that finance, industrial, and property influenced

 54

positively the growth in Phisix. Mining is not statistically significant in the

model and was dropped from the prediction model.

For every unit increase in finance, industrial and property, the Phisix

increased by 0.412, 0.487 and 0.492 index points, respectively. As

previously discussed above, the finance sector provided the necessary

capital stock to the different sectors that propel the daily growth of the money

market (Phisix) and the economy, in the long run. The re-invested capital (in

the form of productive and provident loans) helped the industry to produce

the consumer goods of the economy. These consumer goods were allocated

and distributed in the property sector (households and private sectors).

The average combined unit increase of finance, industrial and

property to Phisix growth is 1.795.44 index points. This means that the three

sectors carry the weight of pushing the Phisix from 1997 – 2005. Without any

daily activities from the different sectors (finance, industry, oil, and property)

and the role of political administration, there is zero (constant = - 88.925)

daily trading transactions in Phisix.

The oil sector negatively affects the daily growth of Phisix. This means

that oil exploration did not attract enough investors within the test period

(1997 – 2005). Oil, as a non-renewable resource, is presently explored in

the high seas of northern Philippines and entails large investment with high

risk attached to it. Quality and volume of local oil extracted were not sufficient
 55

to satisfy the increasing demand for it. Thus, a unit increase in oil investment

pulled down daily Phisix activity within 1997 - 2005. Table 6 shows the

different sectors that affect Phisix.

Table 6. Regression Results for Phisix.

Constant Finance Industry Oil Property GMA

b -88.925 0.412 0.487 -1.809 0.492 6.257
Se b 15.739 0.010 0.003 0.839 0.006 3.382
t -5.650 41.896 163.304 -2.156 78.625 1.850
sig. 0.000 0.000 0.000 0.031 0.000 0.064
D -W = 1.981 R= 0.993 Std error= 3.317
2
ADJ R = 0.986 F= 31983.39

Statistical standard (low standard of error of the b-coefficients, high

t-values, high significance level) for accepting the coefficients (b) for each

determinant (finance, industrial, oil, and property) showed evidence of

significance at 1% and 5%, respectively. The level of significance (0.000)

shows that the study’s chance of being wrong could be 1 out of a thousand

and (.05) 5 out of a hundred.

This means that the combined activities of finance, industrial, and

property sectors pushed up the daily trading activities of Phisix within 1997 –

2005. There is strong multiple correlation (R = 0.993) between the

determinants (finance, industry, oil, property and the effect of GMA

 56

administration) to Phisix. The qualitative variable GMA is significant at 10%

level.

The adjusted R-square (Adj R2 = 0.986) shows proof that the

combined effects of the determinants can ably predict 98.6% of Phisix daily

activities. The predicted Phisix model closely approximates the actual data of

the Phisix as evidenced by low standard error of the estimates (std error =

3.317) for the whole regression and the high F – value (= 31983.39 > 10).

The Prais-Winsten method for correcting autocorrelation of the variables

shows absence of autocorrelation (DW = 1.981 > 1.831).

2. Evaluate the predictability of the Phisix model.

The Phisix model derived for this study is in the form:

Phisix = - 88.925 + 0.412 (F) + 0.487(I) – 1.809 (O) + 0.492(P) +

6.257(GMA)

Where: Phisix = Philippine Stocks Exchange Composite Index

F = Finance sector (index)

I = Industry sector (index)

O = Oil sector (index)

 57

P = Property sector (index)

GMA = is a dummy variable representing the political

influence of the GMA administration within January

21, 2001 – December 2005). That is,

GMA = 1, if the period covered the administration of GMA

0, if the period were those of others’ administration

a = - 88.925

bis = 0.412, 0.487, - 1.809, 0.492 and 6.257, respectively.

Using the reference criteria of Harvey (Gujarati, 1999), the derived

Phisix equation passed the standard for accepting a good model like:

parsimony, identifiability, goodness-of-fit, theoretical consistency, and

predictive power. The Phisix prediction model is parsimonious because it

captures the simplified version of reality. The set of data (2349 days of

trading activities of Phisix) has one identifiable estimate per parameter. The

goodness-of-fit criterion implies that the explanatory variables (finance,

industrial, oil and property) explain 98.6% of the variation in the Phisix. The

signs of the explanatory variables conform to theoretical consistency such

that the positive signs of the 3 sectors and qualitative variable (finance,

industrial, property and the political environment) indicate a push in the daily
 58

trading activities of Phisix. The critical role of the oil sector signifies the high

risk in investing in the exploration of extracting oil in the high seas.

The predicted values of the Phisix model are comparably close to the

actual values of the 2349 daily PSE index. The closeness of the actual and

predicted values indicates a small standard error of estimate (SEE) of about

3.317 index points (see Table 7, Figures 3 and 4) around the means of the

PSE index. The small residual indicates that the difference between the

actual and predicted values of Phisix is close to zero.

The study found that the derived Phisix prediction power approximates

well the actual data. Table 7 shows the predictive power of the Phisix Model.

Table 7 is presented in abbreviated form to minimize on space. The 2349

data set is presented in the appendix. Figure 3 shows the estimated Phisix

model superimposed with the actual observations. Figure 4 exhibits

individual daily pattern of actual and predicted Phisix together with the

residual term. Hence, the Phisix model is acceptable and can be used for

predicting Phisix daily activities.

 59

Table 7. Predictive Power of the Estimated Phisix Model.

Abbreviated actual and predicted values of Phisix (including the residual).

day Phisix predictd resid day Phisix predictd resid day Phisix predictd resid
1 3170.56 3213.68 -43.12 46 3292.30 3293.30 -1.00 91 2694.40 2692.28 2.12
2 3154.48 3170.74 -16.26 47 3279.82 3284.63 -4.81 92 2682.10 2679.16 2.94
3 3169.26 3169.59 -0.33 48 3286.45 3286.27 0.18 93 2671.39 2669.98 1.41
4 3186.37 3190.39 -4.02 49 3289.34 3290.21 -0.87 94 2671.39 2671.57 -0.18
5 3195.47 3192.65 2.82 50 3280.99 3279.19 1.80 95 2677.11 2673.67 3.44
6 3206.99 3203.49 3.50 51 3270.64 3271.27 -0.63 96 2705.92 2707.74 -1.82
7 3223.36 3224.53 -1.17 52 3261.24 3263.27 -2.03 97 2631.09 2639.48 -8.39
8 3262.14 3259.73 2.41 53 3234.79 3232.35 2.44 98 2576.13 2572.05 4.08
9 3270.31 3271.68 -1.37 54 3231.86 3227.52 4.34 99 2509.36 2500.77 8.59
10 3261.72 3263.67 -1.95 55 3215.64 3215.89 -0.25 100 2499.38 2498.46 0.92
11 3258.52 3256.19 2.33 56 3200.61 3198.29 2.32 101 2538.97 2537.13 1.84
12 3290.19 3291.73 -1.54 57 3206.67 3205.77 0.90 102 2586.69 2588.72 -2.03
13 3294.31 3297.43 -3.12 58 3201.65 3202.24 -0.59 103 2598.20 2614.30 -16.10
14 3280.85 3279.78 1.07 59 3222.46 3220.88 1.58 104 2598.20 2598.41 -0.21
15 3279.23 3281.48 -2.25 60 3204.12 3207.80 -3.68 105 2710.91 2695.20 15.71
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
2296 1954.29 1956.57 -2.28 2314 2072.05 2071.55 0.50 2332 2113.60 2112.23 1.37
2297 1944.62 1946.49 -1.87 2315 2063.74 2065.25 -1.51 2333 2100.25 2102.40 -2.15
2298 1922.50 1921.74 0.76 2316 2060.75 2062.45 -1.70 2334 2101.40 2101.03 0.37
2299 1917.47 1916.40 1.07 2317 2056.13 2058.00 -1.87 2335 2099.04 2101.52 -2.48
2300 1927.49 1927.80 -0.31 2318 2079.98 2079.86 0.12 2336 2097.25 2111.77 -14.52
2301 1934.55 1933.86 0.69 2319 2126.56 2130.51 -3.95 2337 2071.67 2069.42 2.25
2302 1941.46 1944.07 -2.61 2320 2130.93 2131.91 -0.98 2338 2047.56 2045.07 2.49
2303 1960.22 1961.03 -0.81 2321 2103.75 2102.51 1.24 2339 2024.70 2022.15 2.55
2304 1960.22 1960.46 -0.24 2322 2119.65 2119.72 -0.07 2340 2062.78 2067.75 -4.97
2305 1960.22 1960.46 -0.24 2323 2106.10 2104.50 1.60 2341 2075.05 2078.91 -3.86
2306 2007.12 2007.63 -0.51 2324 2106.10 2106.39 -0.29 2342 2078.58 2081.68 -3.10
2307 2031.70 2034.73 -3.03 2325 2107.36 2107.50 -0.14 2343 2111.46 2113.73 -2.27
2308 2031.70 2031.95 -0.25 2326 2099.74 2098.96 0.78 2344 2111.46 2111.82 -0.36
2309 2030.05 2027.75 2.30 2327 2086.18 2088.49 -2.31 2345 2088.60 2089.87 -1.27
2310 2046.93 2045.79 1.14 2328 2114.89 2112.74 2.15 2346 2067.32 2061.46 5.86
2311 2097.30 2101.52 -4.22 2329 2116.30 2114.22 2.08 2347 2096.04 2103.13 -7.09
2312 2099.14 2100.83 -1.69 2330 2099.15 2098.72 0.43 2348 2096.04 2096.40 -0.36
2313 2092.68 2095.71 -3.03 2331 2103.42 2104.28 -0.86 2349 2096.04 2096.40 -0.36
 60

Figure 3. Predictive Power of the Phisix Model

4000.00

3500.00

3000.00

2500.00
PSEindex

actual
2000.00
predicted
1500.00

1000.00

500.00

0.00
1 221 441 661 881 1101 1321 1541 1761 1981 2201
day

Figure 4. Actual, Predicted, and Residual of Phisix Model

8000.00

7000.00

6000.00

5000.00
PSE index

4000.00 resid
predicted
3000.00 PSEi
2000.00

1000.00

0.00
1 255 509 763 1017 1271 1525 1779 2033 2287
-1000.00
DAY
 61

3. Determine the effect of political administration to the daily activities

of the Philippine stock market (Phisix).

The Phisix model depicts numerical or quantitative [- 88.925 +

0.412(F) + 0.487(I) –1.809(O) + 0.492(P)] and qualitative [+ 6.257 (GMA)]

parameters. The quantitative variable (finance, industry, oil, and property)

numerically influenced the daily trading activities of the Phisix. However, the

trading activities of the Philippine stock market (proxy by the Phisix) can be

influenced by the system (i.e., political, economic and others) where it exists.

The system effect indicates the presence or absence of qualities or attribute

that affect Phisix trading activities. This qualitative variable is the dummy

variable (GMA) that represents the time frame of Pres. Gloria Macapagal –

Arroyo’s administration.

The dummy variable takes the values 1 or 0 --- 1 representing the

period of GMA administration and 0 representing the absence of her

administration (or others’ administration). As a rule in using a dummy

variable (also known as categorical, nominal, dichotomous), there should be

(m – 1) dummy variables in the model to avoid the situation of perfect

multicollinearity (Gujarati, 1999). This means that only one parameter should

be introduced in the model (2 – 1 = 1) to avoid the dummy variable trap.

 62

If the daily activities of Phisix is affected by a characteristic or political

influence, then the derived Phisix equation (1) cannot be used

straightforwardly for prediction because its structure is not stable (Gujarati,

1999). This means that within the Phisix equation, the daily Phisix activities is

affected, although marginally, by the political system or environment

(statutes, laws, ordinance, government administration, etc.).

Therefore, the Phisix prediction equation

Phisix = - 88.925 + 0.412(F) + 0.487(I) – 1.809(O) + 0.492(P) + 6.257(GMA)

(1)

as affected by the GMA administration takes the following form:

Average Phisix covered by the GMA administration:

A[Phisix |b5 = 1, Xi] = (a + b5) + b1(F) + b2(I) – b3(O) + b4(P)

A[Phisix |b5 = 1, Xi] = - 82.668 + 0.412(F) + 0.487(I) – 1.809(O) + 0.492(P)

(2)
 63

Average Phisix not covered by the GMA administration:

A[Phisix | b5 = 0, Xi] = a + b1(F) + b2(I) – b3(O) + b4(P)

A[Phisix |b5 = 0, Xi] = - 88.925 + 0.412(F) + 0.487(I) – 1.809(O) + 0.492(P)

(3)

Where: A = average

Phisix = Philippine Stock Exchange Composite Index

Xi = any sectors (financial, industrial, oil) that affects Phisix

i = 1, 2, 3, 4.

GMA = dummy variable for Government effect to Phisix

= 1, if the Phisix is under the time period of GMA

= 0, if the Phisix is not under the time period of GMA

Equations 2 and 3 are nonstochastic (values are fixed) and assumes

that the error term (ui) is equal to 0; A(ui = 0). Since the dummy variable

(GMA) takes the values one (1) and zero (0), there were no fluctuations in

the data set that will represent the noise or error term. Hence, the error term

in equations 2 and 3 is equal to zero (0).

 64

The inclusion of the dummy variable GMA showed that political

administration affects the Phisix daily activities within January 1997 –

December 2005. This means that the Phisix varies per given term of political

administration. This variation within the equation is called the “scale effect” or

heteroscedascity. The scale effect showed that the GMA administration

(January 21, 2001 – December 2005) pushed up the Phisix by about 6.267

index points or an average of 1708.71 PSE index points per day. Estimated

daily trading activities of Phisix not covered by the GMA administration

indicate an average 1702.45 PSE index points per day. Table 8 shows the

abbreviated results of the actual predicted Phisix (without GMA and GMA).

Figure 5 exhibits the push of the GMA administration from other

administration within the specified period. Figure 6 shows the application of

the Phisix model in day 2349 (Dec. 31, 2005).

In this case, the fluctuations in the Phisix daily trading activities were

tested for political administration, financial crisis, and technological

obsolescence (time trend). Except for the political effect, the rest of the

qualitative or dummy variables (Asian Financial Crisis and technological

obsolescence) were found to be statistically insignificant in the regression

and were dropped from the equation.

 Table 8. Effect of GMA Administration to Phisix

ABBREVIATED RESULTS FOR ACTUAL, PREDICTED and GMA Administration

day Phisix predicted w/o gma gma day Phisix predicted w/o gma gma day Phisix predicted w/o gma gma
1 3170.56 3213.68 3213.73 3219.99 46 3292.30 3293.30 3372.99 3379.25 91 2694.40 2692.28 2740.60 2746.86
2 3154.48 3170.74 3213.73 3219.99 47 3279.82 3284.63 3364.98 3371.24 92 2682.10 2679.16 2725.17 2731.43
3 3169.26 3169.59 3228.59 3234.85 48 3286.45 3286.27 3371.07 3377.32 93 2671.39 2669.98 2712.84 2719.10
4 3186.37 3190.39 3249.48 3255.74 49 3289.34 3290.21 3374.47 3380.73 94 2671.39 2671.57 2712.84 2719.10
5 3195.47 3192.65 3255.48 3261.73 50 3280.99 3279.19 3363.97 3370.22 95 2677.11 2673.67 2714.96 2721.21
6 3206.99 3203.49 3263.24 3269.50 51 3270.64 3271.27 3353.89 3360.15 96 2705.92 2707.74 2745.44 2751.70
7 3223.36 3224.53 3280.57 3286.82 52 3261.24 3263.27 3346.18 3352.44 97 2631.09 2639.48 2678.84 2685.09
8 3262.14 3259.73 3316.69 3322.95 53 3234.79 3232.35 3316.92 3323.18 98 2576.13 2572.05 2619.60 2625.86
9 3270.31 3271.68 3326.01 3332.26 54 3231.86 3227.52 3309.29 3315.55 99 2509.36 2500.77 2544.05 2550.30
10 3261.72 3263.67 3319.13 3325.39 55 3215.64 3215.89 3293.00 3299.26 100 2499.38 2498.46 2533.01 2539.27
11 3258.52 3256.19 3313.36 3319.62 56 3200.61 3198.29 3275.34 3281.59 101 2538.97 2537.13 2570.62 2576.88
12 3290.19 3291.73 3346.31 3352.57 57 3206.67 3205.77 3280.18 3286.44 102 2586.69 2588.72 2620.24 2626.50
13 3294.31 3297.43 3353.32 3359.58 58 3201.65 3202.24 3275.43 3281.69 103 2598.20 2614.30 2647.71 2653.97
14 3280.85 3279.78 3338.55 3344.80 59 3222.46 3220.88 3294.35 3300.61 104 2598.20 2598.41 2647.71 2653.97
15 3279.23 3281.48 3338.92 3345.18 60 3204.12 3207.80 3279.40 3285.65 105 2710.91 2695.20 2744.50 2750.76
. . . . . . . . . . . . . . .
. . . . . . . . . . . . . . .
. . . . . . . . . . . . . . .
. . . . . . . . . . . . . . .
2301 1934.55 1933.86 1981.79 1988.05 2319 2126.56 2130.51 2190.85 2197.11 2337 2071.67 2069.42 2142.38 2148.64
2302 1941.46 1944.07 1991.09 1997.34 2320 2130.93 2131.91 2195.89 2202.15 2338 2047.56 2045.07 2115.45 2121.71
2303 1960.22 1961.03 2010.41 2016.67 2321 2103.75 2102.51 2167.17 2173.43 2339 2024.70 2022.15 2089.73 2095.98
2304 1960.22 1960.46 2010.41 2016.67 2322 2119.65 2119.72 2182.85 2189.10 2340 2062.78 2067.75 2132.47 2138.73
2305 1960.22 1960.46 2010.41 2016.67 2323 2106.10 2104.50 2167.41 2173.66 2341 2075.05 2078.91 2148.28 2154.54
2306 2007.12 2007.63 2057.59 2063.85 2324 2106.10 2106.39 2167.41 2173.66 2342 2078.58 2081.68 2154.57 2160.82
2307 2031.70 2034.73 2084.96 2091.22 2325 2107.36 2107.50 2168.53 2174.78 2343 2111.46 2113.73 2189.37 2195.63
2308 2031.70 2031.95 2084.96 2091.22 2326 2099.74 2098.96 2159.85 2166.11 2344 2111.46 2111.82 2189.37 2195.63
2309 2030.05 2027.75 2080.76 2087.01 2327 2086.18 2088.49 2148.32 2154.57 2345 2088.60 2089.87 2167.43 2173.69
2310 2046.93 2045.79 2096.26 2102.52 2328 2114.89 2112.74 2174.59 2180.84 2346 2067.32 2061.46 2139.94 2146.19
2311 2097.30 2101.52 2150.62 2156.88 2329 2116.30 2114.22 2173.64 2179.90 2347 2096.04 2103.13 2175.42 2181.67
2312 2099.14 2100.83 2153.89 2160.15 2330 2099.15 2098.72 2155.79 2162.05 2348 2096.04 2096.40 2175.42 2181.67
2313 2092.68 2095.71 2150.20 2156.46 2331 2103.42 2104.28 2160.66 2166.91 2349 2096.04 2096.40 2175.42 2181.67

65
 66

Figure 5. Effect to Phisix Daily Activities of GMA Administration

Figure 6. Application of the Phisix Model to Day 2349 (Dec. 31, 2005)
 67

Chapter VI

CONCLUSIONS AND RECOMMENDATIONS

The results of the analyses done on the data gathered for the

purpose of determining the risk relationship of the business and industry

sectors to the behavior of the Philippine Stock Exchange Index, or the

Phisix, yield a definitive argument for the problem stated at the beginning of

this research paper.

The model developed in this research grew from the defining CAPM

theory which proposes the beta criterion as the measure of market risk. In

this instance, the author utilized the CAPM theory which is hinged on the

assumption of linear relationships between variables. This study showed

that in the case of the Philippine model, the risk behavior of the Philippine

stock market as measured by the Phisix, does not behave linearly. Because

of the use of the SPSS software, however, the author was able to proceed

to test the predictability of the CAPM-derived model to ascertain and derive

sound conclusions. These are as follows:

 68

1. Research and analyses based on critical and fundamental tests for a

good determining model revealed that there is clearly a determining

relationship between the risk behavior of four of the five stock market

sectors with the Philippine Stock Exchange Composite Index. These

are the property, industrial, financial, and oil sectors. These sectors

contribute to the riskiness of the Philippine Stock Exchange in terms

of the market risk common to the entire Philippine business and

industry. The mining sector showed no significant effect. From the

viewpoint of investing such risk behavior, these sectors points to

higher returns potential as the Phisix strengthens. It must be recalled

that the Asian Financial Crisis of 1997 was said to be spawned by

risky investments in the property sector. But while there appears to

be greater risk in the property sector, there is also higher return

expected on such investments.

2. Data analysis showed that the CAPM theory which rested on the

assumption of linearity among the dependent variable and

independent variables does not hold for the Philippine stock market

where the relationships are clearly not linear. However, after

controlling for the behavior of the data in order to meet the test of

predictability, it was shown that the Phisix model and the conclusions
 69

derived from it reflect significant risk behavior between variables.

There is strong evidence to support the contention that indeed the

Phisix reflects the behavior of Philippine business and industry and

not just the best-performing stocks. The Phisix is driven to perform or

under-perform, as it is, by the performance of industry sectors in their

management of the market risk.

It is evident from the results of our analysis that investments in

oil exploration companies are taking away funds that would have

been invested in other sectors. Resources that should be used for oil

exploration in the Philippines are not directed to the stock market.

The high risks in oil exploration are tied to high cost of technology

that often only foreign companies would have the funds and

expertise.

3. Clearly, the effect of business activity in the property, industrial,

financial, and oil sectors on capital formation as represented by the

Philippine Stock Exchange has been shown to be significant. These

sectors have determined for good or bad the rise and fall of the

Philippine stock market for the time duration of this study. It is

expected to continue doing so as well.

 70

Therefore, the validity of the Phisix as the barometer of Philippine

business and industry and not just the best-selected trading performers

should not be doubted.

The implications for the findings and conclusions that have been derived

from this research are focused on some practical recommendations that

may be gleaned in an effort to strengthen Philippine business and industry,

the role of the Philippine stock market as cheap source of funds and future

research that may take off from this study. These are:

1. There is an imperative for Filipino business acumen to drive the

growth of local companies in the property, industrial, and financial

sectors, tame the riskiness of the Philippine stock market, and

contribute to the effort of raising capital for business. These critical

sectors have proved to be determinants of the behavior of the

Philippine stock market. Noting the interrelationships of the industry

sectors where industrial and property companies depend on financial

companies for credit and banking services, it is essential that Filipino

companies are managed well and made competitive, efficient, and

sustainable. The financial health of these companies explain the

robustness of the Philippine stock market. From the viewpoint of

 71

investors, there is reasonably higher returns to be gained from

investments in these publicly-listed companies and there is every

reason to expect that these companies are well-managed in order to

protect investment risk. Therefore, the key is the development of

Filipino management skill and talent that will drive the development

of the Philippine stock market and encourage more investors to

invest and participate in capital formation for Philippine business and

industry.

2. The significant negative effect of the oil industry on the behavior of

the Philippine stock market must be considered in the light of the

failure of the industry to produce successful forays in oil exploration

and generation. Thus, investment in such companies must be

tempered with a reasonable expectation for funnelling such

investment activity in perhaps better equipment and machinery and

better technical expertise and technology that will undertake a major

oil find for the local companies. The undertaking to reverse the

impact of the oil industry on the effort of local capital formation in the

Philippine stock market may well require the effort of the private

sector and the public sector combined. The oil industry must be able

to demonstrate viable commercial activity in oil exploration in the

 72

Philippines. Oil exploration efforts must be directed towards

discovery and generation of oil deposits. Without such a clear and

direct outcome, investments in oil can only be risky speculative

business that contribute to profit-taking trading activities perhaps but

no real investment effort to strengthen local oil companies and fund

their production activities.

To this end, government can make investments in oil exploration

attractive by granting incentives as well as improving access to

credit.

3. The impact of the political climate cannot be underestimated. The

government must continue to provide business and industry and its

stakeholders with the environment that will allow commerce to

prosper. This will require good governance practices and good

government at all levels from national to local to barangay levels.

This will call for the creation of carefully-crafted legislation aimed at

promoting the development of local business and industry. This

appeals to the judiciary to interpret the laws and decide on disputes

fairly and strongly. This requires for the implementation of

government projects to build well maintained infrastructure, whether

 73

in land, air or by sea, or in I.T. and telecommunications, that will

support business growth and activity. This study has clearly shown

that political climate is a factor in the performance of the Philippine

stock market. In particular, the presidency of GMA has proved to

have a positive effect on the movement of the Phisix. While the

political achievements of the GMA presidency is debatable, the

economic indicators have pointed to resounding success in the

management of the fiscal and monetary bottomlines. Under the GMA

presidency, interest rates went down, inflation rate slowed, the peso

strengthened against the US dollar, and the external debt were

managed. It appears that Philippine business and industry in the

Philippine stock market has responded positively to this environment.

But economics can only be supportive. There should be a change of

face of Philippine politics, moving towards a strong social order

where rampant corruption is tamed down as well.

4. On the side of the investors of the Philippine stock market, especially

institutional investors which have the clout to determine the

directions and the health of local companies, there should be an

impetus to compel these companies to adopt the practices called for

in the principles of responsible investment. There should be effort for

 74

these investors to ensure well-run and well-managed companies that

adhere to good corporate governance, corporate social

responsibility, adherence to international law and regulations, and

environmental, social and governance (ESG) guidelines.

5. For small investors, the continued growth and drive of the Philippine

stock market means healthy returns even in the long-run so that if

the companies in the Philippine stock market are well-managed and

adhere to strong corporate governance and responsible investment

practices, then there would be more of small investors who would

balance risk and return in the stock market. At the moment,

managed funds, including mutual funds, are becoming attractive

investment vehicles for even small investors. Stock market

investments remain risky business but with higher risk comes high

returns. Companies in the Philippine stock market must be able to

demonstrate that investing in their company is worth the risk.

Companies must be able to generate wealth for its investors if not in

dividends then in shareholder value. Only with such companies in its

fold will the Philippine stock market succeed in its mission.

 75

6. Future research can be directed to employing other tools of analysis

such as Data Envelopment Analysis in which the effect of managerial

skill on the performance of individual companies can be measured.

This exercise will shed light on the role of Filipino management skill

in explaining the performance of companies in the Philippine Stock

Exchange. Future research can possible look into firm-level data and

draw benchmarks and best practices for Philippine business and

industry.

It is recommended that businesses, the academe, and the Philippine

Stock Exchange itself as well as institutional investors should continue to

support research aimed at analyzing the events that occurred in the past

and not too recent past in order to understand the present and influence or

caution the directions of the future.

 76

BIBLIOGRAPHY

Aquino, Rodolfo Q. (2006). Efficiency of the Philippine Stock Market

[Electronic version]. Applied Economics Letters, vol. 13, no. 7.

Aquino, Rodolfo Q. (2005). Exchange Rate Risk and Philippine Stock

Returns: Before and After the Asian Financial Crisis [Electronic, working
paper version]. Applied Economics Letters, vol. 15, no. 11, 765-771.

Aquino, Rodolfo Q., “A Multifactor Model of Philippine Stock Returns Using

Latent Macro Risk Factors,” Applied Economics Letters, Vol. 11, No. 11,
December 2004, 961-968. (working paper version)

Balaba, Jeaneth Michelle L. (2007). Best Practices of Leading Pension

Funds: Is the GSIS Up To It? Management Development Program Paper.
Pasay City: Government Service Insurance System (Unpublished).

Bautista, Carlos C. (2005). How Volatile are East Asian Stocks During High-
Volatility Periods [Electronic, working paper version]. Applied Economics
Letters, vol. 12, 2005, 319-326.

Bautista, Carlos C. (2003). Stock Market Volatility in the Philippines

[Electronic version]. Applied Economics Letters, vol. 10, 2003, 315-318.

Bradfield, David. (2003). Investment Basics XLVI, On Estimating the Beta

Coefficient (Electronic Version). Investment Analysts Journal, Number 57.

Brigham, Eugene F. and Houston, Joel F. (2004). Fundamentals of

Financial Management (10th edition). Singapore: Southwestern.

Campbell, John Y., Polk, Christopher and Vuolteenaho, Tuomo. (2005).

Growth or Glamour? Fundamentals and Systematic Risk in Stock Returns
(Electronic Version). National Bureau of Economic Research Working
Paper Series, Working Paper 11389. Retrieved 7 October 2006 in
http://www.nber.org/papers/w11389.

Carver, Robert H. and Nash, Jane Gradwohl. (2000). Doing Data Analysis
with SPSS 10.0. United States: Duxbury.
 77

Chari, Anusha and Henry, Peter Blair. (2004). Risk Sharing and Asset
Prices: Evidence from a Natural Experiment (Electronic version). The
Journal of Finance, Volume 59, Number 3.

Constand, Richard. (1999). Types of Risk. Retrieved 7 October 2006 in

http://www.uwf.edu/rconstand/5994content2003/T4-RiskReturn/T4
RiskReturn1.htm.

Corhay, Albert, Hawawini, Gabriel and Michel, Pierre. (1987). Seasonality in

the Risk-Return Relationship: Some International Evidence (Electronic
version). The Journal of Finance, Volume 42, Number 1, pp. 49-68.

De Ocampo, Pedro Jr. B. (2004). Alternative Methodologies for testing

CAPM in the Philippine Equities Market [Electronic version]. Philippine
Management Review, Volume 11, No. 1.

Diacogiannis, George and Makri, Paraskevi. (2008). Estimating Betas in

Thinner Markets: The Case of the Athens Stock Exchange (Electronic
version). International Research Journal of Finance and Economics, Issue
13.

Elton, Edwin J. and Gruber, Martin J. (1994). Modern Portfolio Theory and
Investment Analysis(4th edition). New York: John Wiley & Sons, Inc.

Fama, Eugene F. and French, Kenneth R. (2006). The Value Premium and
the CAPM (Electronic version). The Journal of Finance, Volume 61,
Number 5.

Ferson, Wayne E., Kandel, Shmuel and Stambaugh, Robert F. (1987). Test
of Asset Pricing with Time-Varying Expected Risk Premiums and Market
Betas (Electronic version). The Journal of Finance, Volume 42, Number 2,
pp. 201-220.

Frankle, A.W. and Hawkins, C.A. (1975). Beta Coefficients for Convertible
Bonds (Electronic version). The Journal of Finance, Volume 30, Number 1,
pp. 207-210.

Gujarati, Damodar. (1999). Essentials of Econometrics (2nd edition). United

States: McGraw-Hill.
 78

Guo, Hui and Whitelaw, Robert F. (2006). Uncovering the Risk-Return

Relation in the Stock Market (Electronic version). The Journal of Finance,
Volume 59, Number 3.

Holton, Glyn A. (2006). Defining Risk (Electronic version). Financial

Analysts Journal, Volume 60, Number 6.

Klemkosky, Robert C. and Martin, John D. (1975). The Effect of Market

Risk on Portfolio Diversification (Electronic version). The Journal of
Finance, Volume 30, Number 1, pp. 147-154.

Li, Ming-Yuan Leon and Lin, Hsiou Wei William.(2003). Examining the
Multiple Volatilities and Co-movements as Well as Beta Coefficients of
International Stock Markets (Electronic version).

Maddala, G.S. (2001). Introduction to Econometrics (3rd edition). England:

John Wiley and Sons.

Michailidis, Grigoris, Tsopoglou, Stavros, Papanastasiou, Demetrios and

Mariola, Eleni. (2006). Testing the Capital Asset Pricing Model (CAPM):
The Case of the Emerging Greek Securities Market (Electronic version).
International Research Journal of Finance and Economics, Issue 4, pp. 78-
91.

Odabasi, Attila. (2000). Evidence on the Stationarity of Beta Coefficients:

The Case of Turkey (Electronic version). Istanbul: Bogazici University.

Panetta, Fabio, Angelini, Paolo, Grande, Giuseppe, Levy, Aviram, Perli,

Roberto, Yesin, Pinar, Gerlach, Stefan, Ramaswamy, Srichander and
Scatigna, Michela (BIS Study Group on Financial Market Volatility) (2006).
The Recent Behavior of Financial Market Volatility (Electronic Version).
Bank for International Settlements, BIS Papers, August, Issue No. 29.

Park, Kwang Woo. (2004). Time-Series Analysis of Return and Beta in U.S.
(Electronic version). Journal of Academy of Business and Electronics,
January, Issue 1.

The Philippine Stock Exchange. (2007). History of the PSE Composite

Index. Unpublished Manuscript.
 79

The Philippine Stock Exchange, Inc. (2004). Philippine Corporate

Handbook. Hong Kong: CEIC Data.Com Ltd.

The Philippine Stock Exchange, Inc. (2004). PSE Fact Book 2004. Manila:
Philippine Stock Exchange.

Vides, Rafael (2007). The Return on Swedish Mutual Funds (Electronic

Version). Sweden: Malardens University.
 80

ANNEX A
SUMMARY MATRIX

Statement of the
Results and
Problem/ Methodology Conclusion Recommendation
Discussion
Objectives

Objective No. 2: Future research

PSEi = - 88.925 The model was can be made into
Test the +0.412 (F) able to predict assessing the
predictability of PSEi = α + + 0.487(I) closely the managerial
the regression – 1.809 (O) behavior of the effectiveness of
model β1Industrial i + + 0.492(P) PSEi using the these determining
+ 6.257(GMA) behavior of the Filipino companies
β2Financial i + + 3.317 sectoral indices. and benchmarking
them, using firm-
β3Propertyi + t-values= -5.6, level data.
41.9, (F)
β4Miningi + 163.3, (I)
Objective No. 3: -2.2, (O) The impact of the Government must
β5Oil i + 78.6, (P) political continue to provide
Assess the impact 1.8 (GMA) environment a conducive
of political β6Dummyi + cannot be business
administrations underestimated. environment for the
on the behavior of adjusted r-sq= .986 The PSEi was industry, especially
the PSEi ε observed to private sector, to
f-stat=31983.39 perform better in thrive.
the GMA
dw stat=1.98 administration than Good government
other and good
administrations. governance at all
levels should be in
place.

SUMMARY MATRIX
Statement of the
Results and
Problem/ Methodology Conclusion Recommendation
Discussion
Objectives

Statement of the In general, business The results validated

Problem: industry sectors have the PSEi as indicator
a significant effect on of the stock market in
Do business industry Regression analysis the risk behavior of general and as
sectors influence the and secondary data the Philippine stock benchmark for
behavior of the research market, as measured Philippine business
Philippine stock by the PSEi and industry.
market indicator,
PSEi?

The oil industry should

Objective No. 1: All industry sectors Overall, the industry reverse its impact on
showed significant sector performance the PSEi and make
Determine the critical PSEi = α + influence on the PSEi goes together with stock investments
industry sectors that β1Industriali + except for the mining the movement of the more attractive by
demonstrating
influence the PSEi β2Financiali + sector. PSEi, with the
significant inroads in oil
and assess its effect. β3Propertyi + exception of the oil exploration.
β4Miningi + Four industry sectors industry which
β5Oili + contribute to the showed a negative Government must
β6Dummyi + riskiness of the impact on the PSEi. develop the oil sector.
ε Philippine stock The oil sector also Investors, especially
market, namely: puts the PSEi most institutional investors,
property, industrial, at risk. should compel
financial and oil. Oil companies to adopt
industry has a the principles of
responsible
negative effect .
investment.
Mining showed no
significant effect.
 1000
1500
2000
2500
3000
3500
4000
500
Phisix Performance Across Political Administrations

0
1/1/1997

RAMOS
5/1/1997
9/1/1997
1/1/1998
5/1/1998
9/1/1998
1/1/1999
5/1/1999

ESTRADA
9/1/1999
1/1/2000
5/1/2000

Annex B
9/1/2000
1/1/2001
5/1/2001
9/1/2001
1/1/2002
5/1/2002
9/1/2002
1/1/2003
ARROYO

5/1/2003
9/1/2003
1/1/2004
5/1/2004
9/1/2004
1/1/2005
5/1/2005
9/1/2005
81
 82

ANNEX C
Regression Results for PSE INDEX

MODEL: MOD_2

Split group number: 1 Series length: 2349

No missing data.

Conclusion of estimation phase.

Estimation terminated at iteration number 9 because:
All parameter estimates changed by less than .001

FINAL PARAMETERS:

Estimate of Autocorrelation Coefficient

Rho .99576124
Standard Error of Rho .00190055

Prais-Winsten Estimates

Multiple R .992756
R-Squared .98556447
Adjusted R-Squared .98552749
Standard Error 3.3168356
Durbin-Watson 1.981059

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 5 1759086.5 351817.30

Residuals 2342 25765.3 11.00

Variables in the Equation:

B SEB BETA T SIG T

FINANS .412037 .009835 .14134475 41.89631 .00000000

INDUSTRI .486803 .002981 .65291623 163.30443 .00000000
OIL -1.809382 .839217 -.00559426 -2.15604 .03118236
PROPERTY .492482 .006264 .29608132 78.62542 .00000000
GMA 6.257029 3.381609 .00470295 1.85031 .06439457
CONSTANT -88.924843 15.738549 . -5.65013 .00000000
 83

ANNEX C
Regression Results for PSE Index

The following new variables are being created:

Name Label

FIT_1 Fit for PSEINDEX from AREG, MOD_2

ERR_1 Error for PSEINDEX from AREG, MOD_2
LCL_1 95% LCL for PSEINDEX from AREG, MOD_2
UCL_1 95% UCL for PSEINDEX from AREG, MOD_2
SEP_1 SE of fit for PSEINDEX from AREG, MOD_2

A generalized least-squares method for estimating a regression equation

whose errors follow a first-order autoregressive process. It cannot be used
when a series contains imbedded missing values. Generally, the Prais-
Winsten method is preferable to the Cochrane-Orcutt method. SPSS HELP
OPTION 1998.
 84

ANNEX D
ITERATION OF ESTIMATION MODEL

Regression

Descriptive Statistics

Std.
Mean Deviation N
PSEINDEX 1718.6048 509.0714 2349
FINANS 579.7964 174.7549 2349
INDUSTRI 2496.1374 723.4336 2349
MINING 2009.8624 988.5154 2349
OIL 2.3230 1.4573 2349
PROPERTY 693.2566 265.6802 2349
AFC .1673 .3733 2349
RAMOS .1656 .3718 2349
ERAP .2852 .4516 2349
GMA .5492 .4977 2349
TREND 1175.0000 678.2422 2349
 85

ANNEX D
ITERATION OF ESTIMATION MODEL

M
o R Adjusted Std. Error
d Squ R of the Durbin-
el R are Square Estimate Change Statistics Watson
R Sig.
Square F
Chang df Cha
e F Change 1 df2 nge
1 .983(a) .967 .967 92.6864 .967 68483.992 1 2347 .000
2 .994(b) .989 .989 54.1940 .022 4519.024 1 2346 .000
3 .998(c) .996 .996 30.8236 .008 4907.103 1 2345 .000
4 .999(d) .998 .998 19.8292 .002 3322.299 1 2344 .000
5 .999(e) .999 .999 17.8324 .000 555.334 1 2343 .000
6 .999(f) .999 .999 17.4859 .000 94.775 1 2342 .000
7 .999(g) .999 .999 17.2840 .000 56.036 1 2341 .000
8 .999(h) .999 .999 17.1585 .000 35.368 1 2340 .000
9 .999(i) .999 .999 170516 .000 30.442 1 2339 .000 .053

Model Summary(j)

a Predictors: (Constant), INDUSTRI

b Predictors: (Constant), INDUSTRI, PROPERTY
c Predictors: (Constant), INDUSTRI, PROPERTY, FINANS
d Predictors: (Constant), INDUSTRI, PROPERTY, FINANS, TREND
e Predictors: (Constant), INDUSTRI, PROPERTY, FINANS, TREND, MINING
f Predictors: (Constant), INDUSTRI, PROPERTY, FINANS, TREND, MINING, OIL
g Predictors: (Constant), INDUSTRI, PROPERTY, FINANS, TREND, MINING, OIL, GMA
h Predictors: (Constant), INDUSTRI, PROPERTY, FINANS, TREND, MINING, OIL, GMA, AFC
i Predictors: (Constant), INDUSTRI, PROPERTY, FINANS, TREND, MINING, OIL, GMA, AFC,
ERAP
j Dependent Variable: PSEINDEX
 86

ANNEX D
ITERATION OF ESTIMATION MODEL

Coefficients(a)
Standar
dized
Mod Unstandardized Coefficie 95% Confidence
el Coefficients nts t Sig. Interval for B

Std. Lower Upper

B Error Beta Bound Bound
1 (Constant) -8.550 6.871 -1.244 .213 -22.025 4.924
INDUSTRI .692 .003 .983 261.694 .000 .687 .697
2 (Constant) 56.005 4.131 13.558 .000 47.904 64.105
INDUSTRI .503 .003 .715 156.814 .000 .497 .509
PROPERTY .587 .009 .306 67.224 .000 .570 .604
3 (Constant) 3.089 2.468 1.252 .211 -1.751 7.928
INDUSTRI .447 .002 .635 224.628 .000 .443 .451
PROPERTY .464 .005 .242 88.103 .000 .454 .475
FINANS .479 .007 .164 70.051 .000 .465 .492
4 (Constant) 91.943 2.213 41.548 .000 87.604 96.283
INDUSTRI .480 .001 .683 342.181 .000 .478 .483
PROPERTY .373 .004 .194 99.494 .000 .365 .380
FINANS .385 .005 .132 82.286 .000 .376 .395
TREND -.046 .001 -.061 -57.639 .000 -.047 -.044
5 (Constant) 77.203 2.086 37.009 .000 73.112 81.294
INDUSTRI .491 .001 .698 364.647 .000 .489 .494
PROPERTY .430 .004 .224 103.553 .000 .422 .438
FINANS .360 .004 .124 82.854 .000 .352 .369
TREND -.043 .001 -.058 -59.656 .000 -.045 -.042
MINING -.020 .001 -.040 -23.566 .000 -.022 -.019
6 (Constant) 70.068 2.173 32.247 .000 65.807 74.329
INDUSTRI .485 .001 .689 327.193 .000 .482 .488
PROPERTY .435 .004 .227 105.973 .000 .427 .443
FINANS .396 .006 .136 70.193 .000 .385 .407
TREND -.043 .001 -.058 -61.038 .000 -.045 -.042
MINING -.012 .001 -.023 -9.931 .000 -.014 -.010
OIL -7.649 .786 -.022 -9.735 .000 -9.190 -6.108
7 (Constant) 68.673 2.156 31.854 .000 64.446 72.901
INDUSTRI .473 .002 .673 222.306 .000 .469 .478
PROPERTY .464 .006 .242 82.857 .000 .453 .475
FINANS .407 .006 .140 70.642 .000 .396 .418
TREND -.030 .002 -.040 -15.911 .000 -.034 -.026
MINING -.012 .001 -.023 -9.691 .000 -.014 -.009
OIL -8.832 .793 -.025 -11.144 .000 -10.386 -7.278
GMA -17.389 2.323 -.017 -7.486 .000 -21.944 -12.834
 87

ANNEX D
ITERATION OF ESTIMATION MODEL
8 (Constant) 83.847 3.330 25.178 .000 77.317 90.378
INDUSTRI .478 .002 .679 211.993 .000 .474 .482
PROPERTY .465 .006 .243 83.660 .000 .455 .476
FINANS .373 .008 .128 46.494 .000 .358 .389
TREND -.037 .002 -.050 -16.723 .000 -.042 -.033
MINING -.010 .001 -.020 -8.581 .000 -.013 -.008
OIL -9.421 .793 -.027 -11.880 .000 -10.976 -7.866
GMA -16.127 2.316 -.016 -6.964 .000 -20.668 -11.586
AFC -12.496 2.101 -.009 -5.947 .000 -16.617 -8.376
9 (Constant) 92.527 3.664 25.250 .000 85.342 99.713
INDUSTRI .476 .002 .677 209.872 .000 .472 .480
PROPERTY .460 .006 .240 82.091 .000 .449 .471
FINANS .392 .009 .135 45.252 .000 .375 .409
TREND -.032 .002 -.042 -13.178 .000 -.037 -.027
MINING -.011 .001 -.022 -9.182 .000 -.013 -.009
OIL -10.762 .825 -.031 -13.050 .000 -12.379 -9.145
GMA -32.432 3.746 -.032 -8.659 .000 -39.777 -25.087
AFC -13.717 2.100 -.010 -6.532 .000 -17.835 -9.599
ERAP -12.680 2.298 -.011 -5.517 .000 -17.186 -8.173

a Dependent Variable: PSEINDEX

MODEL: MOD_2
C

Model Description:

Variable: PSEINDEX
Regressors: FINANS
INDUSTRI
MINING
OIL
PROPERTY
AFC
RAMOS
ERAP
GMA
TREND

95.00 percent confidence intervals will be generated.

Split group number: 1 Series length: 2349

No missing data.
 88

ANNEX D
ITERATION OF ESTIMATION MODEL

Termination criteria:
Parameter epsilon: .001
Maximum number of iterations: 10

Initial values:

Estimate of Autocorrelation Coefficient

Rho 0

Prais-Winsten Estimates

Multiple R .99944102
R-Squared .99888235
Adjusted R-Squared .99887805
Standard Error 17.051577
Durbin-Watson .05291958

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 9 607812681.7 67534742.4

Residuals 2339 680078.9 290.8
C

Variables in the Equation:

B SEB BETA T SIG T

FINANS .391864 .0086596 .13451976 45.25202 .0000000

INDUSTRI .476049 .0022683 .67650552 209.87224 .0000000
MINING -.011099 .0012088 -.02155145 -9.18155 .0000000
OIL -10.762193 .8247057 -.03080837 -13.04974 .0000000
PROPERTY .460289 .0056071 .24022129 82.09077 .0000000
AFC -13.716819 2.0998347 -.01005923 -6.53233 .0000000
RAMOS 32.432244 3.7455993 .02368703 8.65876 .0000000
ERAP 19.752423 2.3933080 .01752322 8.25319 .0000000
TREND -.031894 .0024203 -.04249262 -13.17759 .0000000
CONSTANT 60.095097 4.4799645 . 13.41419 .0000000
 89

ANNEX D
ITERATION OF ESTIMATION MODEL

Iteration History:

Iteration Rho SE Rho DW MSE

1 .97308709 .00476575 1.8662933 11.624924

2 .99217557 .00258207 1.9656049 11.047963
3 .99468250 .00212995 1.9757293 11.018499
4 .99510601 .00204358 1.9772781 11.015084
5 .99520481 .00202290 1.9776327 11.014355
6 .99522972 .00201765 1.9777217 11.014175
7 .99523612 .00201630 1.9777446 11.014129
8 .99523778 .00201595 1.9777505 11.014117
9 .99523820 .00201586 1.9777520 11.014114

Conclusion of estimation phase.

Estimation terminated at iteration number 10 because:
All parameter estimates changed by less than .001

FINAL PARAMETERS:

Estimate of Autocorrelation Coefficient

Rho .99523831
Standard Error of Rho .00201584

Prais-Winsten Estimates

Multiple R .99277255
R-Squared .98559735
Adjusted R-Squared .98553574
Standard Error 3.3187517
Durbin-Watson 1.9777524
C

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 9 1762183.0 195798.11

Residuals 2338 25751.0 11.01
 90

ANNEX D
ITERATION OF ESTIMATION MODEL

Variables in the Equation:

B SEB BETA T SIG T

FINANS .410757 .009887 .14090224 41.54488 .00000000

INDUSTRI .486587 .003003 .65268230 162.03852 .00000000
MINING .000842 .001173 .00196682 .71821 .47269717
OIL -2.025786 .865158 -.00626211 -2.34152 .01928885
PROPERTY .491987 .006279 .29572372 78.36002 .00000000
AFC 1.251774 2.349029 .00132331 .53289 .59416034
RAMOS -7.395008 4.759793 -.00554668 -1.55364 .12040554
ERAP -6.856496 3.410504 -.00725533 -2.01041 .04450295
TREND -.020166 .016876 -.00301213 -1.19492 .23224063
CONSTANT -57.070162 25.152145 . -2.26900 .02335922

The following new variables are being created:

Name Label

FIT_4 Fit for PSEINDEX from AREG, MOD_2

ERR_4 Error for PSEINDEX from AREG, MOD_2
LCL_4 95% LCL for PSEINDEX from AREG, MOD_2
UCL_4 95% UCL for PSEINDEX from AREG, MOD_2
SEP_4 SE of fit for PSEINDEX from AREG, MOD_2

AREG
MODEL: MOD_3
C

Model Description:

Variable: PSEINDEX
Regressors: FINANS
INDUSTRI
OIL
PROPERTY
RAMOS
ERAP
GMA
 91

ANNEX D
ITERATION OF ESTIMATION MODEL

95.00 percent confidence intervals will be generated.

Split group number: 1 Series length: 2349

No missing data.

Termination criteria:
Parameter epsilon: .001
Maximum number of iterations: 10

Initial values:

Estimate of Autocorrelation Coefficient

Rho 0

Prais-Winsten Estimates

Multiple R .99936961
R-Squared .99873962
Adjusted R-Squared .99873639
Standard Error 18.096093
Durbin-Watson .05352056

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 6 607725829.3 101287638.2

Residuals 2342 766931.4 327.5

Variables in the Equation:

B SEB BETA T SIG T

FINANS .500069 .0054384 .17166463 91.95204 .0000000

INDUSTRI .443205 .0012238 .62983226 362.14259 .0000000
OIL -18.642311 .6024297 -.05336637 -30.94521 .0000000
PROPERTY .486454 .0048849 .25387645 99.58328 .0000000
RAMOS 21.940514 2.0811072 .01602435 10.54271 .0000000
GMA -47.502974 1.0258493 -.04644024 -46.30599 .0000000
CONSTANT 50.888161 2.2034983 . 23.09426 .0000000
C
 92

ANNEX D
ITERATION OF ESTIMATION MODEL

Iteration History:

Iteration Rho SE Rho DW MSE

1 .97261278 .00480390 1.8335572 11.871355

2 .99383965 .00229059 1.9734974 11.026736
3 .99540803 .00197840 1.9797995 11.009024
4 .99567602 .00191993 1.9807875 11.006756
5 .99573799 .00190616 1.9810124 11.006264
6 .99575326 .00190275 1.9810676 11.006144
7 .99575708 .00190189 1.9810814 11.006114
8 .99575804 .00190168 1.9810848 11.006107

Conclusion of estimation phase.

Estimation terminated at iteration number 9 because:
All parameter estimates changed by less than .001

FINAL PARAMETERS:

Estimate of Autocorrelation Coefficient

Rho .99575828
Standard Error of Rho .00190162

Prais-Winsten Estimates

Multiple R .99275606
R-Squared .9855646
Adjusted R-Squared .98552143
Standard Error 3.317545
Durbin-Watson 1.9810857

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 6 1759103.0 293183.83

Residuals 2341 25765.3 11.01
 93

ANNEX D
ITERATION OF ESTIMATION MODEL

Variables in the Equation:

B SEB BETA T SIG T

FINANS .412038 .009837 .14134528 41.88747 .00000000

INDUSTRI .486803 .002982 .65291597 163.26963 .00000000
OIL -1.809234 .839406 -.00559380 -2.15537 .03123421
PROPERTY .492483 .006265 .29608182 78.60834 .00000000
RAMOS -.193854 3.310347 -.00014543 -.05856 .95330750
GMA 6.255211 3.382406 .00470159 1.84934 .06453503
CONSTANT -88.874754 15.748720 . -5.64330 .00000000
C

The following new variables are being created:

Name Label

FIT_5 Fit for PSEINDEX from AREG, MOD_3

ERR_5 Error for PSEINDEX from AREG, MOD_3
LCL_5 95% LCL for PSEINDEX from AREG, MOD_3
UCL_5 95% UCL for PSEINDEX from AREG, MOD_3
SEP_5 SE of fit for PSEINDEX from AREG, MOD_3

AREG
MODEL: MOD_4

Model Description:

Variable: PSEINDEX
Regressors: FINANS
INDUSTRI
OIL
PROPERTY
ERAP
GMA

95.00 percent confidence intervals will be generated.

 94

ANNEX D
ITERATION OF ESTIMATION MODEL

Split group number: 1 Series length: 2349

No missing data.

Termination criteria:
Parameter epsilon: .001
Maximum number of iterations: 10

Initial values:

Estimate of Autocorrelation Coefficient

Rho 0

Prais-Winsten Estimates

Multiple R .99936961
R-Squared .99873962
Adjusted R-Squared .99873639
Standard Error 18.096093
Durbin-Watson .05352056

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 6 607725829.3 101287638.2

Residuals 2342 766931.4 327.5
C

Variables in the Equation:

B SEB BETA T SIG T

FINANS .500069 .0054384 .17166463 91.95204 .0000000

INDUSTRI .443205 .0012238 .62983226 362.14259 .0000000
OIL -18.642311 .6024297 -.05336637 -30.94521 .0000000
PROPERTY .486454 .0048849 .25387645 99.58328 .0000000
ERAP -21.940514 2.0811072 -.01946437 -10.54271 .0000000
GMA -69.443488 1.8107047 -.06788990 -38.35164 .0000000
CONSTANT 72.828674 2.6287684 . 27.70448 .0000000
 95

ANNEX D
ITERATION OF ESTIMATION MODEL

Iteration History:

Iteration Rho SE Rho DW MSE

1 .97261278 .00480390 1.8335572 11.871355

2 .99383965 .00229059 1.9734974 11.026736
3 .99540803 .00197840 1.9797995 11.009024
4 .99567602 .00191993 1.9807875 11.006756
5 .99573799 .00190616 1.9810124 11.006264
6 .99575326 .00190275 1.9810676 11.006144
7 .99575708 .00190189 1.9810814 11.006114
8 .99575804 .00190168 1.9810848 11.006107

Conclusion of estimation phase.

Estimation terminated at iteration number 9 because:
All parameter estimates changed by less than .001

FINAL PARAMETERS:

Estimate of Autocorrelation Coefficient

Rho .99575828
Standard Error of Rho .00190162

Prais-Winsten Estimates

Multiple R .99275606
R-Squared .9855646
Adjusted R-Squared .98552143
Standard Error 3.317545
Durbin-Watson 1.9810857

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 6 1759103.0 293183.83

Residuals 2341 25765.3 11.01
C
 96

ANNEX D
ITERATION OF ESTIMATION MODEL

Variables in the Equation:

B SEB BETA T SIG T

FINANS .412038 .009837 .14134528 41.88747 .00000000

INDUSTRI .486803 .002982 .65291597 163.26963 .00000000
OIL -1.809234 .839406 -.00559380 -2.15537 .03123421
PROPERTY .492483 .006265 .29608182 78.60834 .00000000
ERAP .193854 3.310347 .00020524 .05856 .95330750
GMA 6.449065 4.714104 .00484730 1.36804 .17143201
CONSTANT -89.068608 15.947495 . -5.58512 .00000000

The following new variables are being created:

Name Label

FIT_6 Fit for PSEINDEX from AREG, MOD_4

ERR_6 Error for PSEINDEX from AREG, MOD_4
LCL_6 95% LCL for PSEINDEX from AREG, MOD_4
UCL_6 95% UCL for PSEINDEX from AREG, MOD_4
SEP_6 SE of fit for PSEINDEX from AREG, MOD_4

AREG

MODEL: MOD_5

Model Description:

Variable: PSEINDEX
Regressors: FINANS
INDUSTRI
OIL
PROPERTY
RAMOS

95.00 percent confidence intervals will be generated.

Split group number: 1 Series length: 2349

No missing data.
 97

ANNEX D
ITERATION OF ESTIMATION MODEL

Termination criteria:
Parameter epsilon: .001
Maximum number of iterations: 10

Initial values:

Estimate of Autocorrelation Coefficient

Rho 0
C

Prais-Winsten Estimates

Multiple R .9987921
R-Squared .99758567
Adjusted R-Squared .99758051
Standard Error 25.040337
Durbin-Watson .03262749

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 5 607023656.4 121404731.3

Residuals 2343 1469104.2 627.0

Variables in the Equation:

B SEB BETA T SIG T

FINANS .611781 .0067444 .21001311 90.70988 .0000000

INDUSTRI .454259 .0016610 .64554055 273.49248 .0000000
OIL -23.222734 .8222938 -.06647851 -28.24141 .0000000
PROPERTY .414329 .0064066 .21623476 64.67198 .0000000
RAMOS 69.443231 2.5055495 .05071817 27.71577 .0000000
CONSTANT -14.785183 2.3335581 . -6.33590 .0000000
 98

ANNEX D
ITERATION OF ESTIMATION MODEL

Iteration History:

Iteration Rho SE Rho DW MSE

1 .98301283 .00379254 1.9135972 11.336077

2 .99448934 .00216633 1.9810863 11.029326
3 .99539065 .00198171 1.9845961 11.020413
4 .99557203 .00194242 1.9852681 11.018914
5 .99561602 .00193277 1.9854294 11.018565
6 .99562716 .00193032 1.9854701 11.018478
7 .99563000 .00192969 1.9854806 11.018455
8 .99563073 .00192953 1.9854832 11.018450
C

Conclusion of estimation phase.

Estimation terminated at iteration number 9 because:
All parameter estimates changed by less than .001

FINAL PARAMETERS:

Estimate of Autocorrelation Coefficient

Rho .99563092
Standard Error of Rho .00192949

Prais-Winsten Estimates

Multiple R .99274774
R-Squared .98554807
Adjusted R-Squared .98551104
Standard Error 3.3194048
Durbin-Watson 1.9854839
Analysis of Variance:

DF Sum of Squares Mean Square

Regression 5 1759783.3 351956.65

Residuals 2342 25805.2 11.02
 99

ANNEX D
ITERATION OF ESTIMATION MODEL

Variables in the Equation:

B SEB BETA T SIG T

FINANS .414178 .009774 .14207828 42.37692 .00000000

INDUSTRI .486700 .002983 .65279034 163.17454 .00000000
OIL -1.768006 .839461 -.00546607 -2.10612 .03529998
PROPERTY .493196 .006257 .29649636 78.82771 .00000000
RAMOS -.253846 3.311587 -.00019042 -.07665 .93890553
CONSTANT -86.877513 15.359907 . -5.65612 .00000000

The following new variables are being created:

Name Label

FIT_7 Fit for PSEINDEX from AREG, MOD_5

ERR_7 Error for PSEINDEX from AREG, MOD_5
LCL_7 95% LCL for PSEINDEX from AREG, MOD_5
UCL_7 95% UCL for PSEINDEX from AREG, MOD_5
SEP_7 SE of fit for PSEINDEX from AREG, MOD_5

AREG

MODEL: MOD_6
C

Model Description:

Variable: PSEINDEX
Regressors: FINANS
INDUSTRI
OIL
PROPERTY
ERAP

95.00 percent confidence intervals will be generated.

 100

ANNEX D
ITERATION OF ESTIMATION MODEL

Split group number: 1 Series length: 2349

No missing data.

Termination criteria:
Parameter epsilon: .001
Maximum number of iterations: 10

Initial values:

Estimate of Autocorrelation Coefficient

Rho 0

Prais-Winsten Estimates

Multiple R .99897351
R-Squared .99794806
Adjusted R-Squared .99794369
Standard Error 23.084656
Durbin-Watson .03026429

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 5 607244172.8 121448834.6

Residuals 2343 1248587.9 532.9

Variables in the Equation:

B SEB BETA T SIG T

FINANS .418101 .0063794 .14352645 65.53963 .00000000

INDUSTRI .433368 .0015265 .61585331 283.88793 .00000000
OIL -11.946586 .7355220 -.03419887 -16.24232 .00000000
PROPERTY .594260 .0050964 .31013926 116.60372 .00000000
ERAP 47.502853 1.3086460 .04214181 36.29924 .00000000
CONSTANT -3.327082 2.1974607 . -1.51406 .13014607
C
 101

ANNEX D
ITERATION OF ESTIMATION MODEL

Iteration History:

Iteration Rho SE Rho DW MSE

1 .98385497 .00369812 1.9174654 11.303038

2 .99459113 .00214628 1.9785885 11.021415
3 .99549524 .00195915 1.9820989 11.012362
4 .99567993 .00191866 1.9827811 11.010822
5 .99572556 .00190852 1.9829478 11.010458
6 .99573734 .00190589 1.9829907 11.010365
7 .99574041 .00190521 1.9830019 11.010341
8 .99574122 .00190503 1.9830048 11.010334

Conclusion of estimation phase.

Estimation terminated at iteration number 9 because:
All parameter estimates changed by less than .001

FINAL PARAMETERS:

Estimate of Autocorrelation Coefficient

Rho .99574143
Standard Error of Rho .00190498

Prais-Winsten Estimates

Multiple R .99275055
R-Squared .98555365
Adjusted R-Squared .98551664
Standard Error 3.3181821
Durbin-Watson 1.9830056

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 5 1759176.2 351835.24

Residuals 2342 25786.2 11.01
 102

ANNEX D
ITERATION OF ESTIMATION MODEL

Variables in the Equation:

B SEB BETA T SIG T

FINANS .413151 .009805 .14172692 42.13696 .00000000

INDUSTRI .486756 .002982 .65285406 163.23382 .00000000
OIL -1.789500 .839429 -.00553275 -2.13181 .03312658
PROPERTY .492840 .006261 .29629431 78.71825 .00000000
ERAP -2.959427 2.375556 -.00313320 -1.24578 .21296879
CONSTANT -85.682749 15.712619 . -5.45312 .00000000
C

The following new variables are being created:

Name Label

FIT_8 Fit for PSEINDEX from AREG, MOD_6

ERR_8 Error for PSEINDEX from AREG, MOD_6
LCL_8 95% LCL for PSEINDEX from AREG, MOD_6
UCL_8 95% UCL for PSEINDEX from AREG, MOD_6
SEP_8 SE of fit for PSEINDEX from AREG, MOD_6

AREG
MODEL: MOD_7

Model Description:

Variable: PSEINDEX
Regressors: FINANS
INDUSTRI
OIL
PROPERTY
GMA

95.00 percent confidence intervals will be generated.

Split group number: 1 Series length: 2349

No missing data.

Termination criteria:
Parameter epsilon: .001
Maximum number of iterations: 10
 103

ANNEX D
ITERATION OF ESTIMATION MODEL

Initial values:

Estimate of Autocorrelation Coefficient

Rho 0

Prais-Winsten Estimates

Multiple R .99933968
R-Squared .9986798
Adjusted R-Squared .99867699
Standard Error 18.516573
Durbin-Watson .04844492

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 5 607689431.6 121537886.3

Residuals 2343 803329.1 342.9
C

Variables in the Equation:

B SEB BETA T SIG T

FINANS .467927 .0046081 .16063078 101.54502 .0000000

INDUSTRI .439612 .0012027 .62472588 365.50984 .0000000
OIL -16.526784 .5812269 -.04731036 -28.43431 .0000000
PROPERTY .520333 .0037645 .27155777 138.21944 .0000000
GMA -52.834146 .9132981 -.05165214 -57.84983 .0000000
CONSTANT 56.653143 2.1841693 . 25.93807 .0000000

Iteration History:

Iteration Rho SE Rho DW MSE

1 .97505577 .00458650 1.8497432 11.728568

2 .99406458 .00224803 1.9743819 11.019077
3 .99544253 .00197055 1.9798849 11.004039
4 .99568609 .00191729 1.9807854 11.001992
5 .99574271 .00190470 1.9809918 11.001543
6 .99575666 .00190158 1.9810424 11.001434
7 .99576015 .00190080 1.9810551 11.001407
8 .99576102 .00190060 1.9810582 11.001400
 104

ANNEX D
ITERATION OF ESTIMATION MODEL

Conclusion of estimation phase.

Estimation terminated at iteration number 9 because:
All parameter estimates changed by less than .001

FINAL PARAMETERS:

Estimate of Autocorrelation Coefficient

Rho .99576124
Standard Error of Rho .00190055

Prais-Winsten Estimates

Multiple R .992756
R-Squared .98556447
Adjusted R-Squared .98552749
Standard Error 3.3168356
Durbin-Watson 1.981059

Analysis of Variance:

DF Sum of Squares Mean Square

Regression 5 1759086.5 351817.30

Residuals 2342 25765.3 11.00
C
Variables in the Equation:

B SEB BETA T SIG T

FINANS .412037 .009835 .14134475 41.89631 .00000000

INDUSTRI .486803 .002981 .65291623 163.30443 .00000000
OIL -1.809382 .839217 -.00559426 -2.15604 .03118236
PROPERTY .492482 .006264 .29608132 78.62542 .00000000
GMA 6.257029 3.381609 .00470295 1.85031 .06439457
CONSTANT -88.924843 15.738549 . -5.65013 .00000000

The following new variables are being created:

Name Label

FIT_9 Fit for PSEINDEX from AREG, MOD_7

ERR_9 Error for PSEINDEX from AREG, MOD_7
LCL_9 95% LCL for PSEINDEX from AREG, MOD_7
UCL_9 95% UCL for PSEINDEX from AREG, MOD_7
SEP_9 SE of fit for PSEINDEX from AREG, MOD_7
 105

ANNEX D
ITERATION OF ESTIMATION MODEL

16000.00

14000.00

12000.00

10000.00 gma
not gma
8000.00
predicted
6000.00 actual

4000.00

2000.00

0.00
1 250 499 748 997 1246 1495 1744 1993 2242
 106

ANNEX E

Sectoral Index Composition, Percent of Sector Companies in Phisix

(1996, partial)

No. Sector Publicly-Listed Companies

1 Financial BPI Union Bank
PCIBank Security Bank
(10 Metrobank Philippine Savings Bank
Stocks) Philippine National Bank Bankard, Inc.
Far East Bank & Trust Co. Vantage Equities, Inc.
% of Sector in Phisix 16.7%
2 Industrial San Miguel Corporation Pilipino Telephone Corp.
Ayala Corporation Benpres Holdings Corp.
(50 First Phil. Holdings Petron Corporation
Stocks) Meralco ABS-CBN Broadcasting
PLDT Aboitiz Equity Ventures
Filinvest Development Corp. Universal Robina Corp.
JG Summit Holdings, Inc. Jollibee Foods Corp.
% of Sector in Phisix 60%
3 Property Ayala Land, Inc. Filinvest Land, Inc.
SM Prime Holdings Robinson’s Land
(10 C&P Homes, Inc. Primetown Property
Stocks) Megaworld Corporation Universal Rightfield
Fil-Estate Land, Inc. Kuok Phil. Properties
% of Sector in Phisix 20%
4 Mining Philex Mining Corp. United Paragon Mining
Lepanto Consolidated
(7 Mining Atlas Consolidated
Stocks) Benguet Corporation Dizon Copper-Silver
Manila Mining
% of Sector in Phisix 3.3%
5 Oil The Philodrill Corp. Trans-Asia Oil & Mineral
(10 South China Petroleum Alcorn Petroleum
Stocks) Oriental Petroleum Vulcan Industrial
% of Sector in Phisix 0%

Source: Philippine Stock Exchange, Notice to Brokers, November 1996

 107

ANNEX E

Sectoral Index Composition, Percent of Sector Companies in Phisix

(1998, partial)

No. Sector Publicly-Listed Companies

1 Financial BPI Union Bank
PCIBank Security Bank
(10 Metrobank Philippine Savings Bank
Stocks) Philippine National Bank Bankard, Inc.
Far East Bank & Trust Co. Vantage Equities, Inc.
% of Sector in Phisix 13.3%
2 Industrial San Miguel Corporation Pilipino Telephone Corp.
Ayala Corporation Benpres Holdings Corp.
(50 First Phil. Holdings Petron Corporation
Stocks) Meralco ABS-CBN Broadcasting
PLDT Aboitiz Equity Ventures
Filinvest Development Corp. Universal Robina Corp.
JG Summit Holdings, Inc. Jollibee Foods Corp.
% of Sector in Phisix 60%
3 Property Ayala Land, Inc. Filinvest Land, Inc.
SM Prime Holdings Empire East, Inc.
(10 C&P Homes, Inc. Primetown Property
Stocks) Megaworld Corporation Universal Rightfield
Fil-Estate Land, Inc. Uniwide Holdings, Inc.
% of Sector in Phisix 20%
4 Mining Philex Mining Corp. United Paragon Mining
(7 Lepanto Consolidated Mining Atlas Consolidated
Stocks) Abra Mining Itogon-Suyoc Mines, Inc.
Manila Mining
% of Sector in Phisix 6.7%
Trans-Asia Oil &
5 Oil The Philodrill Corp. Mineral
(6 South China Petroleum Alcorn Petroleum
Stocks) Oriental Petroleum Vulcan Industrial
% of Sector in Phisix 0%

Source: Philippine Stock Exchange, Memo for Brokers, February 1998

 108

DATA SHEET

JEANETH MICHELLE L. BALABA

Jeaneth

Yellowbell corner Tulip Sts.

Soldiers Hills 4, Molino, Bacoor, Cavite
Residence Tel. No. (046) 5724188/ (02) 4863712
Cell Phone No. (0917) 9598910
Office Tel. No. (02) 8916161 local 4124/4944/ (02) 8321894

e-mail 1: jm_balaba@yahoo.com
e-mail 2: jm.balaba@gmail.com

CAREER OBJECTIVE

To seek career advancement and promote company growth

by providing excellent customer/client-focused service.

EDUCATION

Master in Business Administration (Techno-MBA Program)

De La Salle University-Dasmariñas
Scholastic Achievement Award, SY 2004-2005
2003-present
March 2008

Diploma in Business Administration

Graduate School of Business
De La Salle University-Dasmariñas
March 2006

B.S. in Information Management

Asia Pacific College
Magallanes Village, Makati City
2001-2002
(33 units earned)
 109

DATA SHEET

EDUCATION

Master in Business Administration (Evening Program)

University of the Philippines
Diliman, Quezon City
1993-1995
(8 units earned)

M.S. in Agricultural Economics

Xavier University
College of Agriculture Complex
Cagayan de Oro City
1988-1989
(27 units earned)

B.S. Economics
Graduated Cum Laude, 1988
School of Economics
University of the Philippines
Diliman, Quezon City

Lourdes College
1973-1980
Kinder to High School
Cagayan de Oro City

Lifetime Member, UP Alumni Association

Member, UP Economics Society, 1986-1988
Member, Junior Philippine Economics Society, 1986-1988
Member, UP Kagayha-an, 1984-1988

Civil service eligibility under P.D. 907

conferred under Certificate No. 17013 on March 29, 1989
 110

DATA SHEET

Research Paper

Master of Business Administration Thesis:

The Influence of Selected Industries on the Risk Behavior of the

Philippine Stock Exchange Composite Index
Successfully Defended on January 2008
De La Salle University-Dasmariñas

Management Development Program Business Case:

Best Practices of Leading Pension Funds: Is the GSIS Up To It?

Presented to the MDP Panel on September 2007
Government Service Insurance System

WORK EXPERIENCE

Records Officer III (Special Assignment)

Bids and Awards Committee Secretariat
GSIS, Reclamation Site, Roxas Boulevard, Pasay City
(March 1, 2006 to present)

• Preparation of the Minutes of the Meetings of the PRBAC

• Drafting of reply to office memorandum and other internal and
external communications
• Other administrative tasks like photocopying, telephone calling

Records Officer III (promotion)

General Services Department
GSIS, Reclamation Site, Roxas Boulevard, Pasay City
(March 2001 to present).

• Act as the Chief of Section responsible for the supervision of the

operations of the Terminated Files Section and the Records Center
under the Mails and Records Division; Supervised a team of seven
(7) records personnel.
 111

DATA SHEET

WORK EXPERIENCE
Records Officer II
General Services Department
GSIS, Financial Center, Reclamation Site, Roxas Boulevard, Pasay City
(February 1999 to March 2001).

• Records management of terminated life insurance policies and of the

Records Center holdings;
• Assist the Chief of Section in managing the Terminated Files Section
and Records Center.

Concurrent Acting Section Chief

Terminated Files Section, Mails & Records Division
General Services Department, GSIS
(June 1999 to March 2001).

• Oversee the records management operations of the Terminated

Files Section and the newly-transferred Records Center, and
supervise the activities of the records personnel of the General
Services Department.

Field Research Analyst, Project Basis

Philippine Center for Marine Affairs, Inc. (PhilMar)
PNOC-ERDC Building, Commonwealth Avenue, Quezon City
(August to October 1998).

• Field research and interview on the economic impact analysis of

cabotage laws with respect to the domestic shipping and export
industries;
• Develop framework of analysis for the economic aspect; Prepared
field report on this issue.

Senior Policy Advocacy Officer

Philippine Exporters Confederation (PHILEXPORT)
ITC Complex, Roxas Boulevard cor. Sen. Gil Puyat Ave. Ext., Pasay City
(September 1997 to March 1998).
• Advocacy support through policy writing based on policy agenda
issues;
 112

DATA SHEET

WORK EXPERIENCE

• Attended meetings for the organization and prepared reports on

these meetings; Prepared speeches and publications foreword;

• Responsible for drafting the draft agreements and terms of reference

for the Export Competitiveness Indicators Program launched as
special project with various government institutions.

Medical Representative (Quezon City area 1991 to 1996, Cagayan de

Oro/Iligan area mid-1996) Duncan Pharmaceuticals (Glaxo Wellcome)
2266 Chino Roces Ave. Ext., Makati City
(February 1991 to October 1996)

• Promotion of select ethical pharmaceutical products to select target

doctors in assigned territory;
• Relationship management and business-building of ethical
medicines of GlaxoWellcome in the Philippines;
• Peer training support as Field Preceptor/Facilitator to new medical
representative trainees at the request of the Sales Training Group
from 1992 to 1996 (function of the Senior Medical Representative);
• Peer support in the negotiation of the Collective Bargaining
Agreement in 1995 which was negotiated successfully with the
management group at a ratification rate of close to 95% by company
field employees.

Marketing Assistant
Corporate Banking Department (Account Management Group I)
Boston Bank of the Philippines, 6764 Ayala Avenue, Makati City
(January to November 1990).

• Relationship management by monitoring and servicing of select

corporate accounts assigned to Account Officer in corporate clients'
loan availments and repricing, money market placements and
international banking import and export transactions;
• Underwent rigorous fast-track training program to familiarize with
banking and trust operations prior to employment regularization on
the third month.
 113

DATA SHEET

WORK EXPERIENCE
Development Specialist
Regional Development Coordination Staff
National and Economic Development Authority
Escriva Avenue, Ortigas Complex, Pasig City
(June 1989 to January 1990)

• Provided research support to the Macroeconomics Division in

regional development concerns and topics;
• Coordinated with government agencies in identifying depressed
areas according to poverty rankings and others;
• Responsible for making representations for the release of the bottom
30% preliminary data from the National Statistics Office required for
a scheduled meeting of top government officials.

College Faculty, Part-Time

Department of History and Political Science,
Xavier University, The Ateneo de Cagayan, Cagayan de Oro City
(June 1988 to March 1989).

• Taught Philippine History and the course on Rizal to college-level

students enrolled in various colleges at the Xavier University.

SEMINARS AND TRAINING

o Management Development Program, Class of 2007, GSIS, March-

September 2007
o Seminar on Procurement and Disposal of Government
Properties,GSIS, March 2007
o PhilGEPS Hands-On Training, PhilGEPS, October 5-6, 2006
o Corporate Awareness Seminar, GSIS, July 18, 2006
o Records Management and Document Security, GSIS, September
2005
o National Convention on Social Security, PICC, Manila, February
2004
o GSIS Administration Group Strategic Planning Conference,
November 2003
 114

DATA SHEET
SEMINARS AND TRAINING

o Seminar on GSIS Operations, GSIS, September 2002

o Preparation of Records Management Manual, Laguna, August
2002
o Integrated Records Management in an IT Environment, Baguio
City, May 2002
o Effective Telephone Communication Skills Workshop, GSIS,
November 2001.
o Basic Supervisory Leadership Seminar, GSIS, September
2001.
o MS Word Training Course, GSIS, August 2001.

o New Performance Evaluation System Workshop, GSIS, August

2000.
o Values Orientation Workshop, GSIS, April 2000.
o Total Customer Service Quality Workshop, February 2000.
o GSIS Orientation, GSIS, January 1999.
o Seminar-Workshop on Records Management, conducted by the
Records Management and Archives Office, Manila, April 1999.
o Seminar on Records Disposition, Records Management and
Archives Office, Manila, 19-21 October 1999.

• Management Advancement Program, Duncan Pharmaceuticals,

1995 -1996
• Field Training Preceptors' Briefing, Duncan, Makati City, 1994
• Management Selection Process, Duncan, September 1993
• Echo Seminar on Seven Habits, Duncan, 1994
• Personal Growth and Career Development, Duncan, 1992
• Inductive Training for Medical Representatives, Glaxo Philippines,
February 1991
• Product Training Program for Marketing Assistants (Phase 1 of the
Account Officership Program), Boston Bank of the Philippines,
February to April 1990.
 115

DATA SHEET

AWARDS AND RECOGNITION

• GM Gold Star Award 1992 (Top 3), 136.5% of 1991-92 target,

Duncan
• GM Silver Star Award 1993, 110% of 1992-93 target, Duncan

PERSONAL

Born 20 October 1967 in Cagayan de Oro City

to Jose Manuel Balaba (deceased), lawyer and former Deputy Treasurer of
the Philippines, and Mercedes Labis Penaga Balaba, teacher, both of
Cagayan de Oro and Misamis Oriental
Single
Table Tennis, Badminton, Lawn Tennis, Walking and Jogging
Interest in research and financial analysis
Interest in conducting negotiations
Interest in training and customer service
Proficient in the English and Filipino written and oral language
Working knowledge of Windows MS Office applications
 116

DATA SHEET

REFERENCES

Maria Luisa Dy
Professor
Graduate School of Business
De La Salle University-Dasmariñas
Dasmariñas, Cavite
(046) 4164554 local 3138
(02) 8447832 local 3138
09178346612

Richie Cruz-Angnged
Trust Manager/Housewife
09178838961/029389529

Lea M. Carreon
Engineering Management Department
GSIS
(02) 8916161 local 4476 or 4477
09178042132

Alan N. Tan
Sales Manager
Filinvest Land
09175011292

Liza Mae T. Martinez

Systems Auditor
ChevronTexaco Philippines
09188015043
Office Tel. No. 028411087

Atty. Nora M. Saludares

Senior Vice-President
Government Services Insurance System
Financial Center, Pasay City
Tel. No. (02) 8916161 local 4534

View publication stats