Question

Consider the following least squares specification between test scores (m and the student-teacher ratio: Ý 607.8 + 5.32 in gncome). According to this equation, a 1% noease income is assocated with an increase in test scores of 0 607.8 points O B. 5.32 points ° C. 0.0532 points O D. 53.2 points Reset Selection Mark for Review What's This? Part 12 of 18-0 Question 12 of 18 1.0 Poir Consider the population regression of log earnings Y where Y- In(Eamings as a function of years of schooling (S), age in years AGE), and years of experience (EXP). If FEMALE is a dummy variable that-1 for female gender and O for male gender, what term(s) should we add to this regression model to allow the y-intercept or expected effect of gender to differ between males and females keeping all else the same? O A FEMALE O B. FEMALE EXP ° C. FEMALE and FEMALE·EXP D. All of these answers are correct □ Mark for Review mats This? Part 13 of 18-Q Question 13 of 18 1.0 Poir True or false. We try running two different regression on the same data set. To test whether Y= 0.1 + 0.2X fits the data better than InM·0.2 + 0.4X. we can use the Adjusted R-squared val and choose the model with the highest Adjusted R-squared. C. True False

Answer 1

Answer:-

1) option (c)

Given that Ybar=607.8+5.32ln(income)

if do the derivative for least square estimation then  d/dx(Y) = 5.32/income [ We know d/dx ln(x)= 1/x]

and 1% increase in income will become 0.01

Y=5.32/income = 5.32/0.01 = 0.0532 points (option c)

2) Since male and female is in binary form so in binary linear regression we should take 1 as independent variables and create dummy variables in accordance with the analysis and taken the main variables and the dummy variables as well in the analysis

so the answer will be (c) Female Female*exp

3) Ans: option (a) True as adjusted R square will give the best possible outcome and give goodness of fit.