What is the predicted probability of approval for a Black non-Hispanic female applicant with a co-applicant?

1. Estimate the OLS model to predict loan approval while controlling for Income, Loan2Income, CoApplicant, Black, Asian, Hispanic, and Female. 2. What is the predicted probability of approval for a Black non-Hispanic female applicant with a co-applicant, earning $100 thousand per year and requesting a loan of $200 thousand? 3. Use the “subset” command to limit your analysis to approved loans. The sample size will be 98,127.AP <-subset(NY, Approved==1): Estimate the OLS model to predict loan amount (a continuous variable) while controlling for Income, CoApplicant, Black, Asian, Hispanic, Female. Note that Loan2Income is not a regressor. 4. I claim that (a) blacks are less likely to be approved for a mortgage than other people of similar income and demography but (b) conditional on being approved they actually receive substantially larger loans. Use the statistical results to analyze these two hypotheses. 5. Use the predict command to get the predicted value and the residual for all 98,127 observations. The commands will be similar toAP$Predict <-predict(model)AP$Resid <-resid(model): What is the variance of the prediction? What is the variance of loan amount? 6. Show how the two variances are used to produce the R-squared value of 0.5568? 7. What is the correlation of loan amount and the prediction? 8. Re-estimate your model using a“log-log” specification. Use log(AP$Loan) as the dependent variable and log(AP$Income) as a regresssor. Include the other variables (for Co-Applicant, Black, Asian, Hispanic, and Female) without making any transformations. 9. I claim that for every percentage point increase in income, the loan amount also increases by one percentage point. Economists call this unit elasticity. Use the log-log model to evaluate this hypothesis. 10. What is the interpretation of the Hispanic variablein the log-log regression model? Be specific about the units.                                                                                                                                                                                      

Sample Solution