Program & Batch:
PGDM (2015-17)
Term:
Term 1
Course Name:
Managerical Economics
Name of the faculty:
Dr. V. J. Sebastian
Topic/ Title :
Dependence of GDP on exports and CPI-IW
Group Number:
10
Group Members:
Sl.
Roll No.
Name
1
2
3
4
5
6
7
150103149
150103062
150101079
150102019
150102089
150102099
150103180
Sandip Ghosh
Faraz Zeeshan
Nitish Agarwal
Ankur Tripathi
Shipra
Suman Gon
Sumedha Anand
Regression Analysis
Regression Analysis is an econometric tool to study the effect of certain
specified dependent variables on a specified independent variable.
There are 2 types of regression models;
Bi-variate :
Y = +X+e
Here, Y is the dependent variable while X is the independent variable. We
are studying the effect that X has on Y. is the hypothetical value of Y
when X=0. While represents the slope coefficient that is , the change in
Y per unit change in X. Here, e is the error term.
Multi-variate : Y = + 1 X1 + 2X2 + 3X3 + ..... + nXn + e
Here, Y is the dependent variable while X is the independent variable. We
are studying the effect that X has on Y. is the hypothetical value of Y
when X=0. While 1 , 2.... n represent the partial slope coefficient that
is , the change in Y per unit change in each of the X's while keeping the
other X's constant. Here, e is the error term.
Linear regression makes several key assumptions:
Linear relationship - The regression equation is linear in
parameters & .
No or little multicollinearity - There exists no relationship
between the various independent variables.
No auto-correlation - The error terms at different points of
time are not related to each other .
Homoscedasticity - The variance of the error terms in different
samples is constant.
Regression
Variables Entered/Removeda
Mode
l
Variables
Entered
Variables
Removed
Method
Exports
(Rs.
Crores),
CPI-IWb
. Enter
a. Dependent Variable: GDPFC (Rs.Billion)
b. All requested variables entered.
Model Summary
Mode
l
1
R
.999a
R
Adjusted R
Square
Square
.998
Std. Error of the Estimate
.998
1422.792
a. Predictors: (Constant), Exports (Rs. Crores), CPI-IW
ANOVAa
Model
Sum of
Squares
df
Mean
Square
4225.8
31
Regressi
on
17109024
973.235
85545124
86.618
Residual
36438089.
432
18
2024338.3
02
17145463
062.667
20
Total
Sig.
.000b
a. Dependent Variable: GDPFC (Rs.Billion)
b. Predictors: (Constant), Exports (Rs. Crores), CPI-IW
Coefficientsa
Model
Unstandardized
Coefficients
Std. Error
Standardiz
ed
Coefficient
s
Beta
Sig.
2356.688
9629.601
(Constant)
1
CPI-IW
-4.086
.001
260.375
32.418
.452
8.032
.000
.029
.003
.552
9.812
.000
Exports (Rs.
Crores)
a. Dependent Variable: GDPFC (Rs.Billion)
Regression with DW test
Model Summaryb
Mode
l
.999a
R
Adjusted R
Square
Square
.998
Std. Error
of the
Estimate
.998
Durbin-Watson
1422.792
.884
a. Predictors: (Constant), Exports (Rs. Crores), CPI-IW
b. Dependent Variable: GDPFC (Rs.Billion)
ANOVAa
Model
Sum of
Squares
df
Mean
Square
4225.8
31
Regressi
on
17109024
973.235
85545124
86.618
Residual
36438089.
432
18
2024338.3
02
17145463
062.667
20
Total
a. Dependent Variable: GDPFC (Rs.Billion)
b. Predictors: (Constant), Exports (Rs. Crores), CPI-IW
Coefficientsa
Sig.
.000b
Model
Unstandardized
Coefficients
Std. Error
CPI-IW
Sig.
Beta
2356.688
9629.601
(Constant)
1
Standardiz
ed
Coefficient
s
-4.086
.001
260.375
32.418
.452
8.032
.000
.029
.003
.552
9.812
.000
Exports (Rs.
Crores)
a. Dependent Variable: GDPFC (Rs.Billion)
Residuals Statisticsa
Minimu
m
Maximu
m
Mean
Std.
Deviation
Predicted
Value
6986.34
107395.
83
38155.
29248.098
33
Residual
3609.78
2667.82
6
8
21
.000
1349.779
21
Std. Predicted
Value
-1.066
2.367
.000
1.000
21
Std. Residual
-1.875
2.537
.000
.949
21
a. Dependent Variable: GDPFC (Rs.Billion)
Regression with VIF
Variables Entered/Removeda
Mode
l
Variables
Entered
Variables
Removed
Method
Exports
(Rs.
Crores),
CPI-IWb
. Enter
a. Dependent Variable: GDPFC (Rs.Billion)
b. All requested variables entered.
Model Summaryb
Mode
l
.999a
R
Adjusted R
Square
Square
.998
Std. Error
of the
Estimate
.998
DurbinWatson
1422.792
.884
a. Predictors: (Constant), Exports (Rs. Crores), CPI-IW
b. Dependent Variable: GDPFC (Rs.Billion)
ANOVAa
Model
Sum of
Squares
df
Mean
Square
4225.8
31
Regressi
on
17109024
973.235
85545124
86.618
Residual
36438089.
432
18
2024338.3
02
17145463
062.667
20
Total
a. Dependent Variable: GDPFC (Rs.Billion)
b. Predictors: (Constant), Exports (Rs. Crores), CPI-IW
Sig.
.000b
Coefficientsa
Model
Unstandardized
Coefficients
Std. Error
CPI-IW
Sig.
Beta
2356.688
9629.601
(Constant)
1
Standardiz
ed
Coefficient
s
-4.086
.001
260.375
32.418
.452
8.032
.000
.029
.003
.552
9.812
.000
Exports (Rs.
Crores)
Model
Collinearity Statistics
Tolerance
VIF
(Constant)
1
CPI-IW
.037
26.793
Exports (Rs. Crores)
.037
26.793
a. Dependent Variable: GDPFC (Rs.Billion)
Collinearity Diagnosticsa
Mode Dimensi
l
on
Eigenval
ue
Condition
Index
Variance Proportions
(Consta
nt)
CPI-IW
Exports (Rs.
Crores)
2.713
1.000
.00
.00
.00
.284
3.093
.03
.00
.03
.003
28.341
.97
1.00
.97
a. Dependent Variable: GDPFC (Rs.Billion)
Residuals Statisticsa
Minimu
m
Maximu
m
Mean
Std.
Deviation
Predicted
Value
6986.34
107395.
83
38155.
29248.098
33
Residual
3609.78
2667.82
6
8
21
.000
1349.779
21
Std. Predicted
Value
-1.066
2.367
.000
1.000
21
Std. Residual
-1.875
2.537
.000
.949
21
a. Dependent Variable: GDPFC (Rs.Billion)
Theoretical Explanation:
2 When a country exports goods, it sells them to a foreign market,
that is, to consumers, businesses, or governments in another
country. Those exports bring money into the country, which
increases the exporting nation's GDP. When a country imports
goods, it buys them from foreign producers. The money spent on
imports leaves the economy, and that decreases the importing
nation's GDP.
3 Net exports can be either positive or negative. When exports are
greater than imports, net exports are positive. When exports are
lower than imports, net exports are negative. If a nation exports,
say, $100 billion dollars worth of goods and imports $80 billion, it
has net exports of $20 billion. That amount gets added to the
country's GDP. If a nation exports $80 billion of goods and imports
$100 billion, it has net exports of minus $20 billion, and that amount
is subtracted from the nation's GDP.
The Consumer Price Index for Industrial Workers (CPI-IW) is an economic
indicator used by the government in India to track inflation for a particular
segment of the consumer market. It does this by establishing a baseline for
the purchasing power of industrial workers at a particular point in time and
comparing what the same amount of money can purchase in later years. If
purchasing power decreases, inflation has caused the prices of consumer
goods to rise. The percentage increase in prices over the baseline is
considered the country's rate of inflation.
So if CPI-IW increases, it shows an overall increase in the price level of the
consumer market of the industrial workers. Since GDP is defined as the total
value of all the goods and services produced during a particular period of
time, the increase in price level leads to an increase in the value of the goods
produced which consequently escalates the GDP.
Analysis :
From the obtained results, the regression equation can be
written as:
GDPFC = -9629.601 + (260.375)CPI-IW + (0.029)Exports
Exports are given in Rs. Crores.
The coefficient of CPI-IW is 260.375 which means that a unit
increase in CPI-IW leads to an increase of 260.375 units in GDPFC keeping
exports constant.
The coefficient of exports is 0.029 which means that a one
crore increase in Exports leads to an increase of 0.029 units in GDPFC
keeping CPI-IW constant.
Since the standardised coefficient for exports is .552 while
that for CPI-IW is .452, the effect of change in exports on GDP is greater
than that for change in CPI-IW.
T-test : It is used to test the hypothesis whether a particular coefficient is
significant or not, i.e whether an independent variable has an effect on
the dependent variable or not.
In our regression model, the following are the t-values :
CPI-IW : 8.032 which is greater than 2 and hence highly significant.
Exports : 9.812 which is greater than 2 and hence highly significant.
F-test :
It tests the statistical fit of the regression equation . In other words, it
tests the hypothesis that all the coefficients are simultaneously
insignificant.
Here, the F value is 4225.831 which is much more than 4. Thus, the
statistical fit of the equation is very good and all the variables are not
simultaneously insignificant.
Also, the sig value for F is .000 which is less than .05 and hence we reject
the null hypothesis that all the variables are simultaneously insignificant.
R2 : The value of this indicates how much variation in the
dependent variable is explained by the variation in the independent
variables.
Here, the value is .998 which means that 99.8% of the variation in the
GDP is explained by the variation in CPI-IW and Exports.
Durbin-Watson statistic for Autocorrelation : If the value of the
DW statistic is close to 2, we say there is no autocorrelation.
However, if the valus is less than 2, then there is evidence of positive
autocorrelation. If the value is more than 2, then there is evidence of
negative autocorrelation.
Since our value for the DW statistic is .884, thus there is
evidence for positive autocorrelation.
Variance Inflation Factor (VIF) : This is used to test
Multicollinearity . If this value is greater than 10, the multicollinearity is
high.
In our case, the VIF value is 26.793 and so there is evidence of high
multicollinearity present between Exports and CPI-IW.
