You are on page 1of 6

t-stat > - t-critic

-1.7116 >
R -2.08
0.12242
Square 4

t Stat -1.7116

GrossIncome(per capita) Line Fit Plot


3

2
Population Growth
Population Growth 1 Predicted Population
Growth
0
50000
-1 0 100000

GrossIncome(per capita)

When we Simple linear regress the Population Growth against


Gross Income (per capita) for the year 2015.In above regression t-
stat is greater than t-critical of -2.08 for 21 degree of freedom.
Hence we fail to reject our null hypothesis of no correlation
between Population growth and gross income per capitafor
confidence level of 95%.
R 0.5161
Square 52

t Stat -
4.7330
8

t-stat < -t-critic


-4.73308 <
-2.08

Literacy Rate(%) Line Fit Plot


3

2
Population Growth
Population Growth 1 Predicted Population
Growth
0
0 100 200
-1
Literacy Rate(%)

Similarly by Simple Linear regression of Population growth against


literacy rate we observe that t-stat is less than -2.08 (the critical
value for t-statistic 21 degree of freedom).Hence we reject the
Null Hypothesis of no correlation between Population Growth
against Literacy rate for confidence level of 95%.
t-stat < -t-
R 0.4174
Square 22

t- Stat -
3.8790
1

critic
-3.87901 <-
2.08

Life Expectancy(years) Line Fit Plot


3
2.5
2
Population Growth
1.5
Population Growth Predicted Population
1
Growth
0.5
0
-0.5 60 100
40 80

Life Expectancy(years)

In Simple Linear regression of Population growth against Life


Expectancy we observe from the sample data that t stat <-t-critic.
Hence we reject the Null hypothesis of No correlation between
Population Growth and life expectancy for confidence level of
95%.

R 0.1204
Square 42

t Stat -
1.6957
7

t-stat > -t-


critic
-1.69577>
-2.08

Health Exp(%of GDP) Line Fit Plot


3

2.5

2
Population Growth
1.5
Population Growth Predicted Population
1 Growth

0.5

0
0 5 10 15
-0.5

Health Exp(%of GDP)


In Simple Linear regression of Population growth against Health
Expenditure (% of GDP) we observe that t-stat> -t-critic. Hence
we fail to reject the Null hypothesis of No correlation between
Population growth and Health Expenditure (% of GDP) for
confidence level of 95%.
Multiple Regression analysis
In order to examine the effects of Literacy rate , Life expectancy , Health expenditure and Per
capita gross income on Population growth rate in the World, a database containing the country
wise stat is used. Since this is a Multiple Regression analysis, the following estimated model
was used:

Y= 0 + 1LR + 2LExp + 3HE +4PCI + Et

Dependent variable= Y=Population Growth Rate


Explainers= LR(Literacy rate), LExp(Life expectancy), HE(Health Expenditure),PCI(Per capita
gross Income)
Residual | Error term= Et

0 = Constant expected value for Y if explainers are equal to zero.

1, 2, 3= population regression coefficients- shows how many units the dependent variable
will change if explainer changes by one unit.
EMPIRICAL RESULTS

SUMMARY
OUTPUT

Regression Statistics
0.78548318
Multiple R 6
0.61698383
R Square 5
Adjusted R 0.53186913
Square 1
0.59302419
Standard Error 7
Observations 23

ANOVA
Significa
df SS MS F nce F
10.197 2.5492 7.2488
Regression 4 04 59 51 0.001162
6.3301 0.3516
Residual 18 99 78
16.527
Total 22 24

Standa
rd Lower Upper Lower Upper
Coefficients Error t Stat P-value 95% 95% 95.0% 95.0%
7.30276257 1.7833 4.0949 0.0006 11.049 3.556 11.049
Intercept 5 56 54 8 3.55607 46 07 46
GrossIncome(p 1.49086E- 1.04E- 1.4287 0.1702 3.68E- 3.68E-
er capita) 05 05 16 1 -7E-06 05 -7E-06 05
- - - -
Literacy 0.02420381 0.0098 - 0.0034 0.044 0.0034
Rate(%) 4 84 2.4489 0.0248 -0.04497 4 97 4
Life - - -
Expectancy(ye 0.05896979 0.0287 2.0480 0.0554 0.0015 0.119 0.0015
ars) 1 93 7 28 -0.11946 22 46 22
- - -
Health Exp(%of 0.00342252 0.0740 0.0462 0.9636 0.1520 0.158 0.1520
GDP) 3 12 4 26 -0.15892 7 92 7

Hypothesis Testing using F-statistic


H1:0= 1= 2= 3= 4=0
H1:Not all of them are equal to zero simultaneously.
The value of F statistic which we get for above multiple regression model is
Degree of dreedom(4,18)
F= 7.248851
Fcritical(1 % significance level)=4.58
F>Fcritical thereby we reject the null hypothesis and conclude that not all of
coefficients are simultaneously zero.

You might also like