(i) Run the regression of log(psoda) on prpblck, log(income), and prppov, using only the observations where there are no missing data. The estimated equation is: log(psoda)ˆ = logpsoda^ = + prpblck + log(income) + prppov. True or False: β1ˆ is statistically different from zero at the 5% significance level but is not statistically different from zero at the 1% significance level against the two-sided alternative. True False
(ii) The correlation between log(income) and prppov is approximately , indicating some degree of . The two-sided p-value for testing H0: β2ˆ = 0 is approximately , indicating that β2ˆ is statistically significant at the 5% significance level and statistically significant at the 1% significance level. The two-sided p-value for testing H0: β3ˆ = 0 is approximately , indicating that β3ˆ is statistically significant at the 5% significance level and statistically significant at the 1% significance level.
(iii) Run the regression of log(psoda) on prpblck, log(income), prppov, and log(hseval) using only the observations where there are no missing data. The estimated equation is: log(psoda)ˆ = logpsoda^ = + prpblck + log(income) + prppov + log(hseval). The two-sided p-value for testing H0: β4ˆ = 0 is approximately , indicating that β4ˆ is statistically significant at the 5% significance level and statistically significant at the 1% significance level.
(iv) True or False: After adding the variable log(hseval) into the regression model, both the coefficient on log(income) and the coefficient on prppov are not statistically different from 0 at the 5% significance level. True False
Use the test for H0: β2 = 0, β3 = 0 against H1: β2 ≠ 0 OR β3 ≠ 0 to answer the following questions. The F-statistic is . The p-value is approximately . True or False: The variables log(income) and prppov are jointly significant at the 5% significance level. True False
(v) True or False: The regression model including log(hseval): log(psoda) = β0 + β1 prpblck + β2 log(income) + β3 prppov + β4 log(hseval) + u appears to be less reliable than the regression model excluding log(hseval): log(psoda) = β0 + β1 prpblck + β2 log(income) + β3 prppov + u when it comes to determining whether the racial makeup of a ZIP code influences local fast-food prices. True False
Leave a Reply