question part1 pls

question part1 pls

Problem 2 (Hypothesis Testing) Suppose you have the following regression model: Score2; = Bo + B1 Scorel; + B2POST; + B3 SIZE; + BASIZE; * Scorel; +U; where Scorel; and Score2, are the average grades received by students in school i before and after a new policy that made only those who hold a teaching certification can be teachers. Only a subset of schools are affected by this policy, and POST, is an indicator variable equal to 1 if the school was affected by this policy and 0 if not. SIZE; is equal to the number of students in school i. You have estimated the model using OLS and you obtained o, B1, B2, B3, B4, B5, selo), sel), sel B2), se($), sela), se(Bs), as well as pair-wise correlations corr(Bi, ;). 1. How would you test the hypothesis that Scorel has the same effect at all size levels? Write out the steps. 1 2. How would you compute the change in Score2 associated with a unit change in Scorel at Scorel = 3 and Size = 20? How would you construct confidence interval for this change? 3. How would you test the hypothesis that effect associated with a unit change in Scorel is equal to 0 at Scorel=4.5 and SIZE=20? Problem 2 (Hypothesis Testing) Suppose you have the following regression model: Score2; = Bo + B1 Scorel; + B2POST; + B3 SIZE; + BASIZE; * Scorel; +U; where Scorel; and Score2, are the average grades received by students in school i before and after a new policy that made only those who hold a teaching certification can be teachers. Only a subset of schools are affected by this policy, and POST, is an indicator variable equal to 1 if the school was affected by this policy and 0 if not. SIZE; is equal to the number of students in school i. You have estimated the model using OLS and you obtained o, B1, B2, B3, B4, B5, selo), sel), sel B2), se($), sela), se(Bs), as well as pair-wise correlations corr(Bi, ;). 1. How would you test the hypothesis that Scorel has the same effect at all size levels? Write out the steps. 1 2. How would you compute the change in Score2 associated with a unit change in Scorel at Scorel = 3 and Size = 20? How would you construct confidence interval for this change? 3. How would you test the hypothesis that effect associated with a unit change in Scorel is equal to 0 at Scorel=4.5 and SIZE=20