In: Statistics and Probability
SUMMARY OUTPUT |
|||||||||
Regression Statistics |
|||||||||
Multiple R |
0.039342 |
||||||||
R Square |
0.001548 |
||||||||
Adjusted R Square |
0.000672 |
||||||||
Standard Error |
1.554036 |
||||||||
Observations |
1142 |
||||||||
ANOVA |
|||||||||
df |
SS |
MS |
F |
Significance F |
|||||
Regression |
1 |
4.267905 |
4.267905 |
1.767229 |
0.183991 |
||||
Residual |
1140 |
2753.131 |
2.415027 |
||||||
Total |
1141 |
2757.398 |
|||||||
Coefficients |
Standard Error |
t Stat |
P-value |
Lower 95% |
Upper 95% |
Lower 95.0% |
Upper 95.0% |
||
Intercept |
1.621275 |
0.067736 |
23.93504 |
6.6E-103 |
1.488373 |
1.754177 |
1.488373 |
1.754177 |
|
X Variable 1 |
1.64E-06 |
1.23E-06 |
1.329372 |
0.183991 |
-7.8E-07 |
4.06E-06 |
-7.8E-07 |
4.06E-06 |
(a)
The sample regression equation to estimate the dependent variable income regressed on the independent variable children (x1)
... (1)
From the given summary output
Substituting the parameter estimates in equation (1), the regression equation is given as
... (I)
(b)
The statistical significance of the model is inferred from F- statistic and the corresponding p-value of the F-statistic.
We test the hypothesis
vs
We reject Null hypothesis if p-value of the corresponding statistic is <= alpha (level of significance = (1 - level of confidence))
Here from the output, consider the p-value of the F-statistic = 0.183991, ...(2)
At 95% level of confidence, alpha = 1 - 95% = 1 - (95/100) = 1 - 0.95 = 0.05 .. (3)
From (2) and (3), p-value (0.183991) > alpha (0.05) we do not reject the null hypothesis i.e. the independent variable, children (x1) considered in the model is not statistically significant in explaining the independent variable, income (y)
(c)
The statistical significance of coefficient for the independent variable, income (x1) is inferred from t- statistic and the corresponding p-value of the t-statistic.
We test the hypothesis
vs
We reject Null hypothesis if p-value of the corresponding statistic is <= alpha (level of significance = (1 - level of confidence))
Here from the output, consider the p-value of the t-statistic = 0.183991, ...(2)
At 95% level of confidence, alpha = 1 - 95% = 1 - (95/100) = 1 - 0.95 = 0.05 .. (3)
From (2) and (3), p-value (0.183991) > alpha (0.05) we do not reject the null hypothesis i.e. the independent variable , children (x1) considered in the model is not statistically significant in predicting the independent variable, income (y)
(d) Interpret the coefficient for the independent variable.
From (I), the coefficient of the independent variable, children (x1) is interpreted as, for a unit increase in the independent variable (x1) results in an increase in the average of the dependent variable, income (y) by 1.64E^-06 (= 0.00000164)
(e)
Coefficient of determination (R Square), determines the percentage of the observed variation in income (y) is explained by the model.
From the given output, R Square = 0.001548 (in percentage terms, 0.001548 * 100 = 0.1548%) i.e, only 0.1548% of the total variation in dependent variable income (y) is explained by the independent variable, children (x1)
(f)
Substituting the value of the independent variable, children (x1) = 3 in equation (I)
Hence the estimated value of Income, for 3 children is 1.6213 (approximately)