(a) (5p) Run a regression of the average price on the BP indicator using the data from the time period during the spill (the "if" statement in Stata will be helpful here, and the subset() command in R will be helpful). Weight the regression using the p_obs_time variable. In Stata, we use "aweight" for this. For example, instead of running "reg y x, robust" we run "reg y x [aweight = p_obs_time], robust". Interpret the coefficient on bp. (b) (5p) Run the regression of price on during using only the BP gas stations (and using only the observations from before the spill or during the spill). Weight the regression using the p_obs_time variable. Interpret the coefficient on during. (c) (6p) What are some omitted variables that might vary across gas stations but don't change over time? How could you deal with these? (d) (6p) What are some omitted variables that might change over time but affect all gas stations equally? How could you deal with these? (e) (8p) Run the Differences in Differences regression. You should: i) Include gas station fixed effects, ii) weight the regression by p_obs_time; and iii) cluster the standard errors by zip code. If you are