Question 8 The binary variable to be explained is approve, which is equal to one if a mortgage loan to an individual was approved, zero otherwise. The key explanatory variable is white, a dummy variable equal to one the applicant is white. Consider the following LPM: approve-Bo+B1white+X'8+u where X includes control variables and u is an error term. Suppose that the following condition holds: E(ul white, X)=E(approve X). Under this condition one can conclude that OBi is identified as the causal effect since white is independent of u OBi is identified as the causal effect only if E(ulwhite,X)=0 holds additionally OBi is not identificd as the causal effect Bi is identified as the causal effect Question 9 Cities often want to determine how much additional law enforcement will decrease their murder rates. A simple model with cross-section data at the provincial level to address this question is (1) murdpc-Bo+Bipolpc+Bzincpc+Bapvty+u where murdpc is murders per capita, polpc is number of police officers per capita, incpc is income per capita and pvty is the percent of people in the city that are below a poverty line and u is the error term. Suppose that the supply of polpc is a function of murdpc described by the following equation (2) polpc=So+Simurdpc+Szincpc+83stax+v where stax is the percent of sales tax in the province in which the city is located and v is the error term. Suppose that u and v are uncorrelated and that pvty and stax are valid instrumental variables. The two stage least squarcs cstimation procedure for equation (2) amounts to O none of the other answers are correct