M For this problem you need to copy and run missing.do and missingdta from Canvas This question illustrates some basic data manipulation techniques for dealing with nissing data. Suppose a researcher is estimating the effects of demographic factors on an individual' income using the General Social Survey of 1991. Her research assistant has tried to prepare the data for her to generate the estimates, and her do file includes some hard to understand comments. Your job is to help decipher exactly what was done by the research assistant and to help interpret her results. a} Based on the frequencies from part 1 of the program how prevalent is n'I'ssing data? Does it exist primarily in the D'v'lllncome}. one or more of the IVs, or both? b} In part 2. why do you think her assistant decided to recode the income variable? Why didn't the assistant think MD was being handled correctly in the original coding? c) In part 3. why does the assistant create the PAEDLEZ and MDPAEDJC variables? Why are they coded that way? d} In part In. the assistant con'rnents that \"This regression model will give us an idea of whether or not the MD in PAEEIE is n'issing on a random basis." How do these regression models accon'plish this? What does the coefficient for MDF'AEDLII supposedly tell you? Is this a valid approach, why or why not