DETERMINATION OF STOCK RETURNS USING FIVE

FACTOR CAPM: EVIDENCE FROM PAKISTAN SOCK

EXCHANGE

ABSTRACT

This study empirically analyses the determination of stock returns using Capital Asset Pricing Model
evidence from Pakistan Stock Exchange (PSX). Panel Regression Analysis is used to analyze the
relationship between the Stock Returns and Premium, Size, Book Market value, Investment,
Profitability. The data is taken from the Financial Statement Analysis of Financial Sector published by
the State Bank of Pakistan (SBP) over the period of 2002 to 2017. Empirical results show that there is
positive and statistically significant relation between the stock returns and size, premium, investment,
profitability. But Book to market value has positive and insignificant relation with stock returns. This
study shows that Fama and French five factors are the best determinants of Stock Returns of financial
firms in Pakistan.

KEYWORDS: CAPM, BMV, Premium, Firm Size, Investment

INTRODUCTION

The appearance of new stock markets which are big and globally diversified and from
time to time returns offered by these markets attract the investors/ shareholders and
the researchers of the finance around the world. Many models are explained by the
researchers to select the portfolio in order to bear the risk and return. The capital asset
pricing modal (CAPM) which is developed by the Sharpe (1964), Linter (1965) and
Mossin (1966) is the base of Asset Pricing Theory. Many decades ago with the new
inventions by the authors the CAPM is still widely used. According to Fama and
French (2004) the authors are still using the CAPM because it gives the intuitions
about the risk and return of emerging/new stock markets. Fama and French (2004)
contribute the two factors size and Book to market value to improve the deficiency of
the single CAPM. But there was still incomplete information/knowledge related to
returns on the stock market.
Different models were pioneered by different authors in different years. All type of
models divided into two types: static and dynamic model. The static model include
Sharpe Lintner CAPM (1964), Black Zero beta CAPM (1972), CAPM of Mayers
(1974), CAPM of Breeden (1979), International CAPM by Solnik (1974), Adler and
Dumas (1983), APT by Ross (1976), three factor model of Fama and French (1993),
three moment CAPM by Rubinsten (1973). The five moment CAPM by Fang and Lai
(1997), DIttmar (1999). The dynamic model includes the International CAPM
(Merton 1973), the Consumption model CAPM by Breedan (1979), production based
CAPM by Lucas (1978) and Brock (1979) Investment based CAPM by Cochrame
(1991), Liquidity based CAPM by Acharya and Pedersan (2005), Conditional CAPM
by Jagannathan and Wang (1996) Fama and French (1997) Momentum model and
Fama and French (1993) three factor model.

Fama and French (1993) proposed a new model. The traditionally capital asset
pricing model was used only one variable to describe the return on stocks. But in
1993 Fama and French introduce a three factor model by including two more factors
in the standard CAPM. These two factors included for the two classes small cap and
stock with low B/P.

Rit − R f = β 0 + B1 ( Rm − R f )it + β 2 SMBit + β3 HMLit + ε it

SMB stand for small minus big refer to the market capitalization and HML stands for
high minus low refer to book to market ratio. B s was the determined by the linear
regression may has positive or negative value. Griffin (2002) challanged the Fama
and French (1993) model by showing that model factors were only country specific,
there must be local factors to provide better explaination.

But the three factor model just only the improvement in the three factor model but it
does not explain the anomalies nor the cross sectional variation in return of stock
market.(Fama and French 2004). To overcome this problem Fama and French (2014)
proposed a new model with five factors contribution two more factors in the three
factor model. Premium, Size, BMV, Investment, Profitability are the five factors of
CAPM used to manipulate the risk and Stock returns of the financial firms. This
study checks whether the components of CAPM Five Factor Model determine the
Stock returns of the financial firms in Pakistan? (Fama and French 2014)

The purpose of this study is to find out the determinants of stock returns for financial
firms listed in Pakistan Stock Exchange (PSX) using five factor CAPM consisting on
Premium, Size, BMV, Investment and Profitability.

Following are the objectives of the research:

 To find out the relation between Premium and Stock Returns of Financial
Firms.
 To investigate the relationship between Size (MC) and Stock Returns of
Financial Firms.
 To check the relationship between BMV and Stock Returns of Financial
Firms.
 To find out the role of Profitability of the firms in determining Stock Returns
of Financial Firms.
 To check impact of Investment on Stock Returns of Financial Firms
Different null hypothesis of this study are following as:
 There is no relation between Premium and Stock returns.
 There is no relation between Size and Stock returns.
 There is no relation between BMV and Stock returns.
 There is no relation between Investment and Stock returns.
 There is no relation between Profitability and Stock returns.
LITERATURE REVIEW

Andre Ricardo et al (2017) investigated the CAPM with dynamic beta in the context
of Brazilian stock market. In the study author selected the beta model to access the
behavior of Brazilian equities, several models were tested to choose the goodness of
fit test and significance parameters. On the basis of given data result showed that time
varying CAPM was less conservative rather than based on CAPM with a static beta.

Chen and Kawaguchi (2018) proposed a multifactor asset pricing model under
Markov Regime in the context of Chinese stock market. Chinese stock market had
two regimes, in 1 regime the risk premium was higher as the author considered when
there was a high risk aversion level. Fama and French three factor model provide
positive risk return on the contra side in regime 2 lower premium observed with low
BMV and big size, it showed that good stocks were much risky in bull market.

Agarwal (2014) proved how to obtain the higher return with low volatility in the
context of Bombay stock exchange (BSE) .Author introduced risk weighted alpha
(RWA) to show the stable increasing return with low level of volatility. Top 30
stocks by market capitalization were taken to test the volatility. At end study
concluded that increasing return and lower volatility showed by stock like Hindustan.
RWA and BSE index maintaining the beta same as in the BSE sensex index.

Terregrossa and Eraslan (2016) proposed the relation between return and beta for the
portfolio in the context of Turkish. In previous studies, no systematic relative
described between beta and portfolio return. But author find a systematic relation
among beta and portfolio return by using the model of security market plane (SMP).
Findings of study showed that SMP model generate more accurate expected returns
which lead to portfolio investment opportunities and decisions.

Kim, Park and Suh (2016) proposed the J-shaped relationship between dividend and
value of firm. Top dividend payers firms also had higher value on the contra side
non-dividend payer firms had lower value. But by using the j-shaped relation, study
find significant results over the period of 1962-2010, after controlling the
characteristics of firms such as profitability, growth and size. Dividend theories such
as free cash flow, catering are inadequate to explain the j-shaped relation.

Jiang, Ma and Shi (2016) observed the stock liquidity and dividend payout in the
context of Chinese listed firms from 2000-2014 and find a positive relation between
stock liquidity and dividend payout. But stock liquidity only showed the liquidity
information and increase incentives to pay dividend to shareholders. Author find
positive relation between stock liquidity and dividend payout but only when the
environment opaque. Dividend payout was favorable for those who had low stock
liquidity. Finally concluded that there was several conditions such as government,
non-controlling block holders and manager-shareholders agency conflict that lead to
stock liquidity.

Banerjee et al (2014) checked the four factor model in the context of Indian stock
market. In the Fama and French three factor model fourth factor momentum was
added to check the impact on stock returns market. Size, BMV, Premium and
Momentum were used as independent variables in the study. The results showed that
three factor model of Fama and French had significant impact but the fourth factor
momentum used by researchers was negligible because it had no significant impact in
study.

Agarwalla at al (2014) checked the impact of Fama and French three factor model
and Momentum factor on the Stock Returns in the context of Indian Equities Market.
Data collected from the period of 1994 to 2014 online data library. According to
study the Average Returns on the base of momentum factor was 21.9 %. Portfolio can
make on the percentage of B/M, Size, and Premium. The B/M percentage was
15.3%, size was 0% and premium was 11.5%. The study was better as compare to
other studies because data was collected for current period of time which gave
accurate results.
Shaker and Eligiziry (2014) compared the asset pricing model in the stock of
Egyptian. GRS test applied to compare the five different models of asset pricing.
Data collected from the EGX100.The CAPM, FF3F, CFF, liquidity four factor model,
five factor model (liquidity and momentum included in the three factor). On the base
of given data from 2003 to 2007 six portfolio was made. Size and BE/ME ratio
considered best for the portfolio. The findings of study confirmed that FF3F model
best for average returns from 2003-2007. Fama and French three factor model has
superiority over the CAPM. The Cahart four factor model was not applicable in
Egyptian market. The liquidity model also had deficiency due to turnover. Results of
FF3F more appropriate at compare to all other models.

Maxim (2014) evaluated the CAPM three factor model and APT at the Romanian
capital market. Sample was taken from the period of 2006-2013 of 25 companies
listed in Bucharest Stock Exchange. In the study 6 models was investigated to select
the best for the Romanian Capital Market. The CAPM two factors, downside-CAPM,
3 and 5 factor model of Fama and French and Arbitrage pricing Theory was used.
The results concluded that for the Romanian capital market Fama and French 5 factor
model was best performing model as compare to other ones because it explaining best
the returns. The other model not rejected because there was some limitations in the
Romanian capital market while getting the data such as transaction, volume low, risk
free security not provided and missing data.

Blanco (2012) used the CAPM and Fama and French three factor model for the
portfolio selection. Study used the data of American NYSE market. The selection of
portfolio depends on the size and BMV. Data was taken from the period of 1926 to
2006. Cross Section Regression and Time Series Regression both are conducted in
study. Sample data based on two factors expected returns and Fama and French
model factors. The results showed the significance statistics and R square value. The
sample period also enough to explain that Fama and French three factor model was
best as compare to single CAPM. On the other hand results also depend on how the
portfolio made in further studies.

Mwall and karasneh (2011) used the Fama and French three factor model in the
context of emerging market to explain the variation in stocks from the period of 1999
to 2010. The study checked the assurance of size and book to market value. Time
series data were used. The procedure made at two stages, firstly select the
independent variable and then dependent variables (portfolios). OLS test was used in
study to show the relation between variables. The results showed that size has
strongly positive impact on ASE. Three factor model of Fama and French provide
better results as compare to single factor model of CAPM.

Hou at al (2011) investigated what factors drive Global Stock Returns. The study
used the monthly returns for 27000 stocks from 49 countries. The multifactor model
includes momentum, cash flow to price, and two variables of CAPM size and book to
market value. In the study B/M, C/P, D/P, E/P, L/B and size variable were used.
Finding showed that local and international versions of these multifactor have lower
price as compare to global and emerging market. Market, C/P and momentum factor
gave the lowest model pricing error and lowest chance to rejection model. But there
were many firms’ characteristics which not considered such as liquidity, stock issue
investment and asset growth. On the base of the study authors reject the
characteristics and find reliable results related to global cash flow to price.

Johnson at al (2010) examined the endogenous leverage and expected stock returns.
This paper described about under which conditions leverage had negative relation
with expected stock returns. This paper investigates the previous model of George
and Hwang (2009) asserts the structure of capital by firms in different default risk
case. Variation cannot control the book-to-market ratio, but may lead to desired
relations expected excess return, market debt and book equity examined. Variation in
taxes and firms duration negatively associated with stock returns. Asset
characteristics also considered while making expectations about expected stock
return. There was negative relation between risk and profitability and expected return
with leverage and positive relation with book to market.

METHODS AND MATERIALS

The equation of this study is following as:

Rit =β 0 + B1PREit + β 2 SIZEit + β3 BMVit + β 4 Invit + β5 Pit + ε it , i =1, 2...n

PRE= Premium( Rm − R f )

SIZE= Show the Market Capitalization

BMV= Book to Market Value

Inv= Investment

P= Profitability

ε = Error term

Rit = stock returns

The Stock returns is dependent variable and Premium, Size, Investment, Profitability
, Book to Market Value are independent variables in this study.

BMV

Size Investment

Stock
Premium Profitability
Returns

Fig 1: Study Framework

Stock return is the dependent variable and premium, size, book market value,
investment and profitability are the independent variables use in this study. For the
calculation of stock returns yearly amount of dividend value is use. To calculate the
value of premium (PRE), the risk free return is minus from the return on market. T
bills value is use for the risk free return. For the calculation of size (market
capitalization) multiplying the market price per share with outstanding shares. Book
to market value (BMV) calculates dividing the share price and book value per share.
Investment (Inv) is the total assets at the end of the year. Profitability (P) value gets
by the value of earning per share (EPS).

Panel data has been cross sectional unit surveyed over time. In panel data there is
space as well as dimension. The other name of panel data is pooled data (pooling of
cross sectional and time series). In this study we use the panel data in a broad logic to
include one or more of these terms and it is called Regression Model based on panel
data regression models. Fixed effect and Hausman test use in this study. Panel data
relate the firms, states, countries, individuals etc. the procedure of panel data consider
the individual specific variable. Panel data by combining the time series and cross
sectional is more informative, more reliable and more effective. Panel data can better
perceive the effect of variables which not correctly measure by alone cross sectional
and time series data.

DATA AND DATA SOURCES

This study use the determinants of stock returns by considering the variables
premium, size, book market value, investment and profitability in Pakistan. The
annual data for the period of 2002 to 2017 is used in this study. Forty two financial
firms are selected as a sample in this study. Firms are selected under the sector of
commercial banks, modaraba, investment banks, leasing companies and insurance
companies. The numbers of commercial banks, modaraba, investment banks,
insurance companies, leasing companies are respectively as 15, 3, 9, 12 and 3. Data is
taken from the Financial Statement Analysis of Financial Sector published by the
State Bank of Pakistan (SBP). All selected financial firms are listed in Pakistan Stock
Exchange (PSX).
RESULTS AND DISCUSSION

This study finds the determinants of stock returns by using capital asset pricing model
in Pakistan. To solve the purpose, the study investigates the Panel Regression
Analysis to check the data. Descriptive statistics, fixed effect model, random effect
model Hausman test use in analysis.

Descriptive statistics

STOCK PREMIUM SIZE BMV INV PROF

RETURNS RISK(PRE)
Mean 1.545972 -15.61891 22.02471 55.11849 157.4196 4.0209133
Median 0.835000 -0.215833 5.759745 24.48077 20.19275 2.120500
Maximum 16.00000 155.0700 341.6893 620.3181 1867.003 104.4800
Minimum 0.0000 -7310.207 0.000275 -139.7436 0.003152 -4.268000
Std.dev 2.213553 353.9591 45.00568 84.51412 284.7112 8.431131
Jarque-bera 1945.196 3182242 5440.332 3282.856 2132.683 75169.33
Probability 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
obervation 428 428 428 428 428 428
Table shows the Descriptive statistics of stock returns, premium, size, BMV,
investment (Inv) and profitability (prof).
The mean value of stock returns is 1.545972 which shows mean value has more
difference from maximum value as compare to minimum value; it means mostly
firms are giving minimum or low dividend to the shareholders. Median value is also
support this argument. Maximum value is 16 which lie in HBL for the year of 2014
and 2015. Minimum value is 0.0000 which lie in Askari bank, BOP, UNICAP, Trust
investment, Jubilee, East West etc for different years. Median value is 0.83 which lie
in CYAN and Habib Metropolitan in the year of 2006 and 2010 respectively. Stock
returns JB value is 1945.196 and probability value less than alpha it means stock
returns is not normally distributed and distribution is skewed ( small values are
more).
The mean value of premium is -15.61891 which show premium risk is high; all firms
are not risk free. Median value is also support this argument. Medina value is -
0.215833 lie in First Habib Modaraba and Capital asset leas in the year of 2013 and
2012, it is the average of both firm values. Maximum value is 155.07 lie in EFU
insurance in the year of 2007. Minimum value is -7310.207 which lie in UNICAP
modaraba 2010. JB value of premium is 3182242 and its probability value less than
alpha it means premium is not normally distributed and distribution is skewed ( small
values are more).

The mean value of size is 22.02471 which show that firm size has variations; small
firms are more as compare to larger firms. 50% firms have more than 50 million
worth. Median value is also support this argument. Median value of size is 5.75945
lie in Adamjee and Jubilee in the year of 2011 and 2012, it is average of both firm
value in the given years. Maximum value of size is 341.6893 lie in MCB 2014.
Minimum value is 0.000275 lie in security investment 2011. JB value of size is
5440.332 and its probability value 0.00000 which less than alpha it means size is not
normally distributed and distribution is skewed (small firms are more).

The mean value of book to market is 55.11849 which closer to the minimum value it
means mostly firms shares and their book value has no difference, there is moderate
difference between share price and book value. Median value is also support this
argument. Median value is 24.48077 lie in East West and EFU in the year of 2008
and 2011 it is average of both firms. Maximum value is 620.31813 lie in JS 2008.
Minimum value is -139.7455 lie in UNICAP 2010. JB value of BMV is 3282.856and
its probability value less tan alpha it means BMV is not normally distributed and
distribution is skewed ( small values are more).
The mean value of investment is 157.4196 which is closer to the maximum value it
means mostly firms are trying to utilize their investment in best way to improve their
profitability. Median value is also support this argument. Median value is 20.19275
lie in Habib Metropolitan and Soneri bank in 2000 and 2001 respectively it is average
of both firm. Maximum value is 1867003389 lie in HBL 2014. Minimum value is
3152 lie in UNICAP 2010. JB value of investment is 2132.683 and its probability
value less than alpha it means investment is not normally distributed and distribution
is skewed.

The mean value of profitability is 4.0209133 which is closer to the minimum value
show that mostly firms are not profitable but they have moderate profit for their
financial firms. Median value is also support this argument. Median value is
2.120500 lie in Bank al habib and STD mod 2008 and 2005 respectively it is average
of both firms. Maximum value is 104.48 lie in EFU general insurance 2008.
Minimum value is -4.268 lie in Askari bank 2013. JB value of profitability
is75169.33 and its probability value less than alpha it means profitability is not
normally distributed.
Correlated Random Effect-Hausman test
Simple regression model is an analysis between the dependent and independent
variable. But it just shows the linear relationship if there is nonlinear relationship then
there is problem to use this model. For the analysis Random Effect model and fixed
Effect model use. The results of Random Effect model are given in
Table1
TEST SUMMARY CHI.SQUARE CHI.SQ D.F PROBABILITY
STATISTICS
CROSS SECTION 56.151015 5 0.0000
RANDOM
Hausman test statistic p- value less than the alpha value so we reject the null
hypothesis and concluded that fixed models are preferable.
Fixed effect model
Results of fixed effect model given in table

Table 2

VARIABLES COEFFICIENT STD.ERROR T-STATISTICS PROB

PRE -0.006829 0.116522 0.602791 0.0376**
SIZE 0.017689 0.002507 7.057019 0.0000***
BMV 0.000646 0.001044 0.618247 0.5368
INVESTMENT 0.005487 0.000552 9.941367 0.0000***
PROFITABILITY 0.019032 0.007769 2.449806 0.0147**
C 0.070238 0.116522 0.602791 0.5470
**,*** show the 5% and 1% level of significance respectively

R-squared 0.761596 Mean dependent 1.545975

var
Adjusted R-squared 0.732813 S.D. dependent var 2.213553
S.E. of regression 1.144190 Akaike info 3.210573
criterion
Sum squared resid 498.7938 Schwarz criterion 3.656318
Log likelihood -640.0626 Hannan-Quinn 3.386617
criter.
F-statistic 26.45932 Durbin-Watson 1.098075
stat
Prob(F-statistic) 0.000000

In table 2 the value of R-square 0.76 it means seventy six percent variation in
dependent variable stock returns is explained through in dependent variables
investment, size, BMV, premium and profitability and remaining 24% is due to other
factors which are not considered in our Regression Model.
F-stat based on F-stat and correspondence p-value we can reject the null hypothesis
(Regression model not significant). We concluded that our model significant at level
of 10%.

On the basis t-stat, the value of probability for coefficient of premium is less than the
one percent so we reject the null hypothesis and concluded that there is positive and
significant relationship between stock returns and premium.

On the basis t-stat, the value of probability for coefficient of size is less than the one
percent so we reject the null hypothesis and concluded that there is positive and
significant relationship between stock returns and size.

On the basis t-stat, the value of probability for coefficient of BMV is not less than the
one percent so we accept the null and concluded that there is positive and
insignificant relationship between stock returns and BMV.

On the basis t-stat, the value of probability for coefficient of investment is less than
the one percent so we reject the null hypothesis and concluded that there is positive
and significant relationship between stock returns and investment.

.On the basis t-stat, the value of probability for coefficient of profitability is less than
the one percent so we reject the null hypothesis and concluded that there is positive
and significant relationship between stock returns and profitability.

CONCLUSION

This study is aimed to find out the determinants of stock returns in the context of
Pakistan. Panel data Regression analysis is used to analyze the stock returns by using
model of Fama and French (2014). Five factors model of Fama and French (2014)
include Premium, Size, Book market value, Investment and Profitability to determine
the stock returns. Data is taken over the period of 2002 to 2017. Forty two financial
firms annually data use in this study to analyze the stock returns. Hausman test
statistic probability value less than the ten percent so we reject the null hypothesis
and concluded that fixed models are preferable. On the basis t-stat, the value of
probability for coefficient of premium is less than the one percent so we reject the
null hypothesis and concluded that there is positive and significant relationship
between stock returns and premium, the value of probability for coefficient of size is
less than the one percent so we reject the null hypothesis and concluded that there is
positive and significant relationship between stock returns and size, the value of
probability for coefficient of BMV is not less than the one percent so we accept the
null and concluded that there is positive and insignificant relationship between stock
returns and BMV, the value of probability for coefficient of investment is less than
the one percent so we reject the null hypothesis and concluded that there is positive
and significant relationship between stock returns and investment, the value of
probability for coefficient of profitability is less than the one percent so we reject the
null hypothesis and concluded that there is positive and significant relationship
between stock returns and profitability. The results indicate that size,premiums,
BMV, investment, profitability are the major determinants of the stock returns in
Pakistan. Seventy six percent variations in the stock returns in Pakistan are due to
premium, size, book market value, profitability, investment. Our results also confirm
that Fama and French (2014) model factors are the major determinants of stock
returns in Pakistan.

REFERENCES

Agarwalla, S. K., Jacob, J., & Varma, J. R. (2014). Four factor model in Indian equities
market

Agarwal, N. (2014). How to obtain high returns with lower volatility in emerging
markets?. Cogent Economics & Finance, 2(1), 890060.

Bedoui, R., & BenMabrouk, H. (2017). CAPM with various utility functions: Theoretical
developments and application to international data. Cogent Economics & Finance, 5(1),
1343230.
Black, F. (1972). Capital market equilibrium with restricted borrowing. The Journal of
Business, 45(3), 444-455.

Black, F. (1974). International capital market equilibrium with investment barriers. Journal of
Financial Economics, 1(4), 337-352.

Breeden, D. T. (1979). An intertemporal asset pricing model with stochastic consumption and
investment opportunities. Journal of financial Economics, 7(3), 265-296.

Banerjee, A., Bandyopadhyay, G., De, A., & Ramani, L. (2014, January). A Study on Carhart
four-factor model in the perspective of Indian market. In Business and Information
Management (ICBIM), 2014 2nd International Conference on (pp. 141-143). IEEE.

Blanco, B. (2012). The use of CAPM and Fama and French Three Factor Model: portfolios
selection.

Chen, J., & Kawaguchi, Y. (2018). Multi-Factor Asset-Pricing Models under Markov Regime
Switches: Evidence from the Chinese Stock Market. International Journal of Financial
Studies, 6(2), 54.

Fama, E. F., & French, K. R. (1993). Common risk factors in the returns on stocks and
bonds. Journal of financial economics, 33(1), 3-56.
Fama, E. F., & French, K. R. (1997). Industry costs of equity. Journal of financial
economics, 43(2), 153-193.

Fama, E. F., & French, K. R. (2004). The capital asset pricing model: Theory and
evidence. The Journal of Economic Perspectives, 18(3), 25-46.

Fama, E. F., & French, K. R. (2014). a five-factor asset pricing model. Journal of Financial
Economics, 123(3), 441-463.

Hou, K., Karolyi, G. A., & Kho, B. C. (2011). What factors drive global stock
returns?. Review of Financial Studies, 24(8), 2527-2574.

Hansen, L. P., & Singleton, K. J. (1983). Stochastic consumption, risk aversion, and the
temporal behavior of asset returns. Journal of political economy, 91(2), 249-265.

Johnson, T. C., Chebonenko, T., Cunha, I., D’Almeida, F., & Spencer, X. (2011).
Endogenous leverage and expected stock returns. Finance Research Letters, 8(3),
132-145.

Liu, L. X., Whited, T. M., & Zhang, L. (2009). Investment-based expected stock
returns. Journal of Political Economy, 117(6), 1105-1139.
Lintner, J. (1965). The valuation of risk assets and the selection of risky investments in stock
portfolios and capital budgets. The review of economics and statistics, 13-37.

Markowitz, H., (1952)"Portfolio selection" The Journal of Finance , Vol, 7. No. 1, pp. 77-91.

Mayers, D. (1974). Portfolio theory, job choice and the equilibrium structure of expected
wages. Journal of Financial Economics, 1(1), 23-42.

Mossin, J., (1966), “Equilibrium in a capital asset market" Econometrics, Vol. 34, No. 4, pp.
768-783.

Ross, S. A. (1976). The arbitrage theory of capital asset pricing. Journal of economic
theory, 13(3), 341-360.

Ronzani, A.R.P.; Candido, O.; Maldonado, W.F.L. Goodness-of-Fit versus Significance: A

CAPM Selection with Dynamic Betas Applied to the Brazilian Stock Market. Int. J. Financial
Stud. 2017, 5, 33.

Sun, F., Liu, C., & Zhou, X. (2017). Analysis of industry risk premium with MVS three
dimensions vector factor model. Cogent Economics & Finance, 5(1), 1374814.

Sharpe, W. F. (1964). Capital asset prices: A theory of market equilibrium under conditions of
risk. The journal of finance, 19(3), 425-442.

Shaker, M. A., & Elgiziry, K. (2014). Comparisons of asset pricing models in the Egyptian
stock market. Accounting and Finance Research, 3(4), 24.

Terregrossa, S. J., & Eraslan, V. (2016). An analysis of the relation between return and beta
for portfolios of Turkish equities. Cogent Economics & Finance, 4(1), 1168501.