Using the data Growth.dta, run three regressions where the dependent variable is Growth and the regressors are:
Regression A: TradeShare and Y earsSchool
Regression B: TradeShare and ln(Y earsSchool)
Regression C: TradeShare, ln(Y earsSchool), Rev Coup, Assassinations, and ln(RGDP60)
Regression D: TradeShare, TradeShare2, TradeShare3, ln(Y earsSchool), Rev Coup, Assassinations, and ln(RGDP60)
Assume that the error term is heteroskedastic.
1. Construct a table to report the regressions above, which includes the estimates, the standard errors, and adjusted R-square.
2. Find the estimated coefficient of ln(Y earsSchool) in Regression B. What is the interpretation of it?
3. In Regression B, test if ln(Y earsSchool) is statistically signi?cant at the 1% level.
4. In Regression C, ?nd the predicted value of Growth when TradeShare is 1.3, Y earsSchool is 8, Rev Coup is .1, Assassinations is .5, RGDP60 is 2800.
5. In Regression D, ?nd the (expected) partial effect of TradeShare on Growth when TradeShare increases from 0.5 to 0.6.
6. In Regression D, is there any evidence of linear relationship between TradeShare and Growth (holding other regressors constant) at the 5% level of signi?cance? To answer this question, test if TradeShare2 and TradeShare3 are jointly statistically signi?cant?
7. Which Regression C or D is a better model to predict Growth? Explain.
Download:- Data.xlsb