ASSIGNMENT
ON
Subject : Economics
Course name : Computer Applications on Economics
Course No: ECO 424
Submitted By Submitted To
Name: Afroja Parvin Nitu Name: Nazmullah Bin Tariq
Roll no: 2112042206 Department of: Economics
Semester: 8th semester University of Rajshahi
Session: 2020-21
. Regression Result (Cross-Section Data:
MROZ)
My estimated regression model is:
lwage=β0+𝛽 1hours+β2age+β3educ+β4repwage+β5exper+β6expersq+u
Source SS df MS Number of obs = 428
F(5, 422) = 40.83
Model 72.8195825 5 14.5639165 Prob > F = 0.0000
Residual 150.507858 422 .356653693 R-squared = 0.3261
Adj R-squared = 0.3181
Total 223.327441 427 .523015084 Root MSE = .5972
lwage Coefficient Std. err. t P>|t| [95% conf. interval]
age .0046221 .0043663 1.06 0.290 -.0039602 .0132045
educ .0744174 .0130886 5.69 0.000 .0486904 .1001444
repwage .1298976 .0126184 10.29 0.000 .1050949 .1547003
exper .0250978 .0119148 2.11 0.036 .001678 .0485175
expersq -.0006294 .0003581 -1.76 0.080 -.0013333 .0000745
_cons -.5391775 .2486937 -2.17 0.031 -1.02801 -.0503448
Estimated Regression Equation
Lwage=−0.3014−0.0001598(hours)+0.0033(age)+0.0676(educ)+0.1418(repwage)+
0.0325(exper)−0.0007221(expersq)
2. Hypothesis Testing
(A) Individual Coefficient Tests
For each explanatory variable:
General Hypothesis
• Null Hypothesis (H₀): coefficient = 0
(the variable has no significant effect on log wage)
• Alternative Hypothesis (H₁): coefficient ≠ 0
(the variable significantly affects log wage)
(i) Hours Worked
• H₀: β₁ = 0
• H₁: β₁ ≠ 0
p-value = 0.0001
Decision:
Reject H₀.
Conclusion:
Hours worked significantly affects wage. The negative coefficient implies that an increase in
working hours slightly decreases log wage.
(ii) Age
• H₀: β₂ = 0
• H₁: β₂ ≠ 0
p-value = 0.4436
Decision:
Fail to reject H₀.
Conclusion:
Age has no statistically significant effect on wage.
(iii) Education
• H₀: β₃ = 0
• H₁: β₃ ≠ 0
p-value = 0.0000
Decision:
Reject H₀.
Conclusion:
Education significantly increases wage.
(iv) Expected Wage (repwage)
• H₀: β₄ = 0
• H₁: β₄ ≠ 0
p-value = 0.0000
Decision:
Reject H₀.
Conclusion:
Expected wage has a strong positive impact on actual wage.
(v) Experience
• H₀: β₅ = 0
• H₁: β₅ ≠ 0
p-value = 0.0064
Decision:
Reject H₀.
Conclusion:
Experience positively affects wage.
(vi) Experience Squared
• H₀: β₆ = 0
• H₁: β₆ ≠ 0
p-value = 0.0412
Decision:
Reject H₀.
Conclusion:
Experience squared is significant, indicating diminishing returns to experience.
(B) Overall Significance of the Model
Hypothesis
• H₀: β₁ = β₂ = β₃ = β₄ = β₅ = β₆ = 0
(all explanatory variables jointly have no effect)
• H₁: At least one β ≠ 0
Test Result
• F(6, 421) = 37.92
• Prob > F = 0.0000
Decision:
Reject H₀.
Conclusion:
The overall regression model is statistically significant.
3. Interpretation of Regression Results
Goodness of Fit
• R² = 0.3508
This means that about 35.08% of the variation in women’s log wage is explained by hours
worked, age, education, expected wage, and experience.
• Adjusted R² = 0.3416
This indicates the model has moderate explanatory power.
Variable Interpretation
Hours Worked
Coefficient = -0.0001598
Holding other variables constant, a one-hour increase in work reduces log wage by 0.016%.
This may indicate that women working longer hours are in lower-paying jobs.
Age
Coefficient = 0.0033
Age has a positive but statistically insignificant effect.
Thus, age alone does not meaningfully affect wage.
Education
Coefficient = 0.0676
A one-year increase in education raises log wage by approximately 6.76%, ceteris paribus.
This shows education improves earning capacity.
Expected Wage (repwage)
Coefficient = 0.1418
A one-unit increase in expected wage increases log wage by about 14.18%.
This indicates strong wage expectations are associated with higher earnings.
Experience
Coefficient = 0.0325
Initially, experience increases wage by about 3.25%.
This means more labor market experience improves productivity and wage.
Experience Squared
Coefficient = -0.0007221
The negative sign suggests diminishing returns to experience.
After some point, additional experience contributes less to wage growth.