Please answer a b c

Free-Response #2. To test whether a mortgage applicant's payment to income ratio affects whether or not a mortgage application is denied, Tobin set up the following regression equation with a binary dependent variable: deny, = Bo + Bi x pi ratio: + U, with the following definitions of independent and dependent variables: Variable Description deny = 1if mortgage application is denied pl_ratio anticipated monthly loan payments/monthly income Tobin obtained the following regression results: Linear regression Number of obs - F1 2378) - Prob > F R-squared Root MSE 2380 37.56 0.0000 0.0397 -31828 Robust Std. Err. deny Coef. P>It! 195Conf. Intervall pi ratio cons 6035349 -.0799096 0984826 0319666 6.13 -2.50 0.000 0.012 4104144 -.1425949 .7966555 -.0172243 (a) (3 points) Interpret the regression results for the pi_ratio variable. What is the model's prediction for an applicant with a pi_ratio=1? (b) (3 points) What is the model's prediction for an applicant with a pi_ratio=2? Discuss any potential issues with this result. (0) (3 points) Tobin re-estimated the relationship using a Probit regression and obtained the below results. Interpret the regression results for the pi_ratio variable. Calculate the change in probability of an increase in piratio from 0.1 to 0.2. Probit regression Number of obs LR chi2 1) Prob > chi2 Pseudo R2 2380 80.59 0.0000 0.0462 Log likelihood - -831.79234 deny Coet. Sud. Err. pi_ratio cons 1951 Cont. Intervall 2.264073 3.67174 -2.446974 -1.941343 2.967907 -2.194159 3591054 .12899 8.26 -17.01 0.000 0.000 Free-Response #2. To test whether a mortgage applicant's payment to income ratio affects whether or not a mortgage application is denied, Tobin set up the following regression equation with a binary dependent variable: deny, = Bo + Bi x pi ratio: + U, with the following definitions of independent and dependent variables: Variable Description deny = 1if mortgage application is denied pl_ratio anticipated monthly loan payments/monthly income Tobin obtained the following regression results: Linear regression Number of obs - F1 2378) - Prob > F R-squared Root MSE 2380 37.56 0.0000 0.0397 -31828 Robust Std. Err. deny Coef. P>It! 195Conf. Intervall pi ratio cons 6035349 -.0799096 0984826 0319666 6.13 -2.50 0.000 0.012 4104144 -.1425949 .7966555 -.0172243 (a) (3 points) Interpret the regression results for the pi_ratio variable. What is the model's prediction for an applicant with a pi_ratio=1? (b) (3 points) What is the model's prediction for an applicant with a pi_ratio=2? Discuss any potential issues with this result. (0) (3 points) Tobin re-estimated the relationship using a Probit regression and obtained the below results. Interpret the regression results for the pi_ratio variable. Calculate the change in probability of an increase in piratio from 0.1 to 0.2. Probit regression Number of obs LR chi2 1) Prob > chi2 Pseudo R2 2380 80.59 0.0000 0.0462 Log likelihood - -831.79234 deny Coet. Sud. Err. pi_ratio cons 1951 Cont. Intervall 2.264073 3.67174 -2.446974 -1.941343 2.967907 -2.194159 3591054 .12899 8.26 -17.01 0.000 0.000