Question

3. Suppose you are trying to estimate the impact of doctor’s visits on health.

(a) You attempt to estimate in the following way Health = ?0 + ?1Doctor’s visits + E ( E= disturbance term) Where the doctor’s visits variable is the number of visits in the last 12 months. Do you believe this estimate will be an accurate measurement of the effect of an additional doctor’s visit on health? If not will it be an over estimate or an underestimate, use math to explain your answer.

(b) Suppose you had panel data, and observed each individual in your data’s age, number of doctors visits and health status for the last 12 years. Would a fixed effects regression eliminate deal with the bias you articulated in the previous question? Why or why not?

(c) Suppose the government provides free medical clinics, but they are only available in a few locations. Some people are lucky enough to be near one, but some people do not live near any. Suppose you know how far away each person in your data lives to the closest medical clinic. How could you use this information to try to estimate the impact of Doctor’s visits on health?

Answer 1