Don't copy the ans.

1. The empirical value of compensating wage differentials may be estimated using average data at the occupational level, rather than individual level data. This may then be applied to, for example, data on average yearly salary and the proportion of workers killed on the job per year for roughly 500 narrowly dened occupations in the Philippines. The following regression results are then obtained: F 20,000 1,000Pl [500) [200] where 1 is the average annual salary for occupation i and P; is the fraction of workers in occupation i that are fatally injured while on the job in a given year. Standard errors are shown in parentheses. a.) Do the results obtained lend credence to the theory of compensating differentials? Why or why not? b.) Are you confident rejecting the theory of compensating differentials on the basis of these results? Why or why not