e) Also include time xed effects in the regression. What might these effects control for? Do your results change as a result of this? 2) In this exercise you are going to evaluate the impact of the construction of a household waste incinerator on the property prices of houses close to the site of the incinerator. You will work with the dataset hprice.dta which contains some of the data used in Kiel and McClain {1995) "House Prices During Siting Decision Stages: The Case of an Incinerator from Rumor Through Operation", Journal of Environmental Economics and Management The dataset contains sale prices of houses sold in North Andorrer1 Massachusetts, in 1978 and 1981. Towards the end of 1978 rumors emerged that a waste incinerator would be build in northwest Andover. In 1981 the contract with the operating company was signed and the plant started operating in 1985. The dataset hprice.dta contains the following variables: price Nominal sale price rprice Real price (deated to 1978 prices) dist Distance to incinerator in feet year = 1978, 1981 insts Distance to interstate in feet rooms Number of rooms area Area of house in square feet land Area of lot in Square feet baths Number of bathrooms cbd Distance to the central business district in feet Answering the follounng questions requires a little digging around in Stata, but moat of your time should be spent thinking about the logic and intuition behind what you're doing. a) A colleague suggests that one can use a regression of the logarithm of the sales prices of houses sold in 1981 on a dummy variable which is equal to one if the house was within three miles of the site of the incinerator to evaluate the impact of the incinerator on housing prices. What is the implicit assumption about the identity of the treatment and control group in this suggestion? What regression estimates do you obtain if you implement this suggestion? What can you conclude from this regression? b) To investigate further the likely causal impact of the incinerator on house prices you decide to estimate the same regression for the data on sales during 1978. What do the results of this regression suggest? c) Now use a differenceindifferences approach to determine the impact of the incinerator on the price of houses close to the incinerator. Under which circumstances will this regression identify the causal effect of the incinerator on housing prices? Do you nd a signith effect of the incinerator on house prices? d) Include a number of control variables in the regression from the previous question: The age of the house, the age squared, the logarithm of the distance to the nearest interstate, the logarithm of the lot size, the logarithm of the area of the house, the number of rooms, the number of bathrooms and the logarithm of the distance to the central business district. How does the inclusion of these variables change your estimates? Why is it sensible to include such control variables in the regression in this application? e) What conclusions would you draw from the regression results for future siting decisions of incinerators