Addis Ababa University
College of business and economics
Department of economic
Econometrics for finance
Test (2): maximum points (10 points)
Name: ___________________________Id. No: _____________ Section: ____________
Given the following regression results, answer the next questions.
. reg lncon i.educ sex_hh age_hh
Source SS df MS Number of obs = 6,767
F(6, 6760) = 162.94
Model 457.933165 6 76.3221942 Prob > F = 0.0000
Residual 3166.4861 6,760 .468415103 R-squared = 0.1263
Adj R-squared = 0.1256
Total 3624.41926 6,766 .535681239 Root MSE = .68441
lncon Coefficient Std. err. t P>|t| [95% conf. interval]
educ
Primary_School .2447878 .0213594 11.46 0.000 .2029166 .286659
Secondary_School .4696264 .0277752 16.91 0.000 .4151783 .5240745
College_educ .5639771 .0343628 16.41 0.000 .4966152 .6313391
Tertiary_educ .7479756 .033579 22.28 0.000 .6821502 .813801
sex_hh -.1924603 .0182436 -10.55 0.000 -.2282235 -.1566971
age_hh .0042669 .0005999 7.11 0.000 .0030909 .0054429
_cons 10.41362 .0329663 315.89 0.000 10.34899 10.47824
Where: lncon is log of consumption expenditure, variables under education refers primary
school complete, secondary school complete, college education, and first degree and above
complete. Sex_hh is gender of household head and age_hh is age of the household head.
A. What is the total number of sample size?
B. What is the 𝑅 2 ? Interoperate it?
C. What is RSS, ESS and TSS?
D. Using the S.E test, which of the explanatory variables are significant and which of them
are not? Show the necessary steps?
E. Using the p-value test approach, identify statistically significant covariates.
F. Interpret the coefficient of educational variables.
2. The result for omitted variable test is given as follow:
Ramsey RESET test for omitted variables
Omitted: Powers of fitted values of lncon
H0: Model has no omitted variables
F(3, 6757) = 11.81
Prob > F = 0.0000
A. Does the module face omitted variables bias problems? Why?
3. The heteroskedasticity test result is given bellow:
Breusch–Pagan/Cook–Weisberg test for heteroskedasticity
Assumption: Normal error terms
Variable: Fitted values of lncon
H0: Constant variance
chi2(1) = 63.85
Prob > chi2 = 0.0000
A. Does the module face heteroskedasticity problems? Why?