Use the data in cps91.dta for this question. These data are for married women. nwifeinc is the woman's nonwage income (i.e. other sources of income, which includes husband's earnings) measured in thousands of dollars; educ and exper are the woman's years of education and prior experience, respectively; kidslt6 is the number of children less than six years old. We also have information on each husband's income and demographics.

Estimate the model

husearns = β0 + β1huseduc +β2husage + β3hushrs +β4husuion + u by OLS and report the results in equation form. Also report both the usual and heteroskedasticity-robust standard errors in parentheses and brackets, respectively. Which of the control variables are statistically significant at the 5% level against the two-sided alternative? Does it matter which standard errors are used?

 

2. reg husearns huseduc husage hushrs husunion Source df MS Number of obs 4,148 F(4, 4143) Prob > F 315.30 Model 120857344 4 30214335.9 0.0000 Residual 397014393 4,143 95827.7559 R-squared Adj R-squared Root MSE 0.2334 0.2326 Total 517871736 4,147 124878.644 309.56 husearns Coef. Std. Err. P>|t| [95% Conf. Interval] huseduc 51.19378 1.69862 30.14 0.e00 47.86357 54.52399 .4719599 0. e00 4.976783 husage hushrs husunion 5.902078 12.51 6.827373 4.060441 .3602757 11.27 1.88 0.e00 0.060 3.354107 4.766775 21.60946 -.8780091 11.47006 33.21344 44.09693 _cons -480.6715 -14.47 0.e00 -545.7876 -415.5553