what determines whether someone wins the lottery?.
a. Would an Show more Q. Suppose you have data on lottery winners and lottery losers and their health status. a. Would an OLS regression of healthi = 0 + 1lottery + ui satisfy the necessary conditions for the OLS estimator to be unbiased? healthi is a measure of health (such as whether or not they eat vegetables daily exercise time alcohol consumption etc.) and lotteryi is a variable which takes on 1 if you win the lottery and 0 if you lose it. Why or why not (hint: what determines whether someone wins the lottery)? b. If the outcome variable (yi) is the number of daily hours devoted to exercise how would you interpret the coeffcient 1 ? Suppose it is estimated to be 1.3. By how much will lottery winners exercise time increase by? Is it statistically significant at the 95 percent level if the estimated standard error on 1 is .48? c. We know the OLS estimator can be written as 1 = (yi-y)(xi-x) / (yi-y)(xi-x) where y(y bar) average of yi and x is the average of xi. Your friend studying public health says they dont have to worry about econometrics they just take the difference of sample means for the treated units (for instance those that win the lottery) and the untreated units (those that dont win). Is there any difference between the OLS estimator and the difference of the sample means in this case and if so why? Show less