(e) [4 points] Which class and crew category coefficients are statistically significantly different at the 1% level from the base case of first-class passengers? Are any of these in the positive direction, or are they all in the negative direction? (f) [4 points] Run the same regression using probit. Copy and paste evidence of your work below. (g) [4 points] The average age is 30.5 and the median age is 29.0. 78%% of the people on board are male. Use the probit model. Focus on a male of age 30. Find the predicted probability of survival for each of the seven classes. Problem 3 [44 points] In this problem we will study how effective campaign expenditures are in congressional elections. To do this, we will use the dataset "votel.dta".' This is a cross-sectional dataset of different election campaigns and how much party "A" and party "B" spent in them. The main variables we will use are the share of votes won by party A voteA, expenditure by party A (in thousands of dollars) expendA, expenditure by party B (thousands of dollars) expendB, the share of voters in the district who voted for party A in the presidential election prtystrA, and a dummy indicating whether party A is the Democratic party democA. (a) [4 points] Generate a dummy variable indicating whether party A won, call it winA. Generate a variable for the difference between expenditure by party A and party B called dSpend. Start by running a linear probability model of the win dummy on expenditures by party A, dSpend, prtystrA and democA. (a.i.) [4 points] What is the interpretation of the coefficient on dSpend? How does the predicted probability of party A winning change when dSpend goes from -250 to -249? How about when prtystrA goes from 0 to 1? Does this make sense? (a.ii.) [4 points]Draw a scatterplot of the actual wins (winA, on the vertical axis) against the predicted winning probabilities. 1 Incidentally, you can access this, and all the other datasets used in the Stock & Watson (https://fmwww.bc.edu/ec-p/data/stockwatson/datasets.list.html) and Wooldridge (http://fmwww.bc.edu/ec-p/data/wooldridge/datasets.list.html) textbooks through Boston College. They even set up a nice stata command that lets you read them straight into stata called bcuse. To install it type "ssc install bcuse" into stata. Then you can load this dataset by typing "bcuse votel, clear"