Professional Documents
Culture Documents
Akshay Kalson
Country
GDP PC($) Inflation(%) Investment Ratio(%) Life Expectancy(years) Literacy Rates(%)
(PCI) (I) (INV) (LE) (LIT)
India 2188 2.83 33.256 68.947 69
China 10099 2.3 43.334 76.386 95
Sri Lanka 4470 4.3 27.853 75.408 91
USA 65062 1.9 20.238 81.097 86
UK 42036 1.9 17.156 81.167 99
Russia 11461 5.3 18.836 70.443 100
Pakistan 1456 9.41 15.118 66.7 55
Bangladesh
Australia
Canada
The data that we took ?
1882
57204
48601
5.55
1.8
1.9
28.972
26.1
23.711
72.709
82.906
82.563
60
99
99
France 43500 1.1 21.476 82.74 99
Germany 49692 1.3 18.828 81.449 99
Japan 41418 0.2 21.998 83.995 99
Kenya 1991 4.35 22.545 63.062 72
Korea
India
USA
Australia
Argentina
Indonesia
Canada
China
Sri Lanka
Russia
France
Zimbabwe
UK
Bangladesh
Kenya
Germany
Japan
Norway
Thailand
Pakistan
0 20
10000
10
0
0
We are going to do LS
What type of analysis we are
doing here?
Explained Variable Explanatory Variable
Inflation
Per capita GDP Investment Ratio
Life Expectancy
Literacy Rate
Let us explain the explanatory
variables first.
Inflation is rise in general level of
prices over time.
Inflation
Economists believeHigh
What Causes high growth
rates ofof
money supply causes high rates of
Inflation?
inflation.
Sometimes fluctuations in inflation are
caused by change in demand of goods or
services available in supply.
Inflation reduces the level of business
investment and hence affects GDP.
Harms of Inflation?
Rising GDP per capita is directly
linked toLife
higher life expectancy .
Expectancy
1$ 33 Days
In our analysis we have taken Investment Ratio as
percentage of GDP.
Investment Ratio
Its lower in poor countries.
Developing countries having low per capita income
have high investment ratios.
Rich countries have higher literacy rates. And hence
Literacy
directly Rate
affect per capita GDP
Regression Results
The independent variables inflation, investment, life
Inference?
and literacy are significant, as can be seen from their t-
statistics and related probabilities.
High adjusted R Square represents good fit!
Som
Soumya
Jarque Bera Test
• Test of normality for a large sample
Jarque Bera Test
• Null Hypothesis states that the Error
terms/residuals follow Normal Probability
distribution
• Test statistic follows Chi Square distribution
(follows for large n) with 2 degree of
freedom
• When the error is normally distributed, this
tests for the joint null hypothesis that
skewness is zero and the kurtosis is 3
Life
Investment Expectancy(yea
GDP PC($) Inflation(%) Ratio(%) rs) Literacy Rates(%)
Country
Mean -1723.462
3 Median -3901.319
Maximum 14377.48
Minimum -18246.12
2
Std. Dev. 8918.463
Skewness 0.187964
1
Kurtosis 2.382700
Jarque-Bera 0.348254
0 Probability 0.840190
-20000 -10000 0 10000
handa
WHITE HETEROSKEDASTICITY TEST
Heteroskedasticity refers to the situation in which the
variance of the error term in the regression equation is
not constant but varies with the independent variable. In
the presence of Heteroskedasticity, the Ordinary Least
Square estimates, although still unbiased are no longer
efficient.
We refer to the WHITE HETEROSKEDASTICITY TEST for the
detection of Heteroskedasticity, wherein one simply
computes an auxiliary regression of the squared OLS
residuals on a constant and all non redundant variables in
the set consisting of the regressors, their squares and
their cross products.
WHITE HETEROSKEDASTICITY TEST
RESULTS
• We see the test statistic nR2 = 14.7639< χ () = . So the
null hypothesis of homoscedasticity cannot be rejected.
Therefore there is no heteroscedasticity in the data.
• Also the p-value corresponding to the f-statistic suggests
that we can not reject the null hypothesis that there is no
heteroskedasticity in the data.
nikhil
Sorted Data
Country
We run two separate regressions corresponding to the first and the last 6 observations and note
the values of RSS thus obtained
Thus the F-statistic obtained is less than the tabular value at both 5% and 1% level of
significance. We therefore accept the null hypothesis of Homoscedasticity (constant variance)
at both 5% and 1% level of significance, i.e., there is no problem of Heteroscedasticity in our
data set
Parameter Stability test
10.0 1.6
7.5
1.2
5.0
2.5
0.8
0.0
-2.5 0.4
-5.0
0.0
-7.5
-10.0 -0.4
6 7 8 9 10 11 12 13 14 15 16 6 7 8 9 10 11 12 13 14 15 16
Since the Cumulative sum is inside the area Here we can say that the parameters
between the two critical lines we can say are constant in terms of variance.
that the parameter are constant in terms of
intercept.
rajan
Serial Correlation Test
• Serial between a serial and its lag value i.e.
the error for the period y is correlated
with the error for the period x
Consequences
• Smaller estimated standard error. As a result,
Larger t value
𝑥ҧ −0
t=
𝑠/
𝑛