2. Use the data in VOTE1 for this exercise, which contains data on election outcomes and campaign
expenditures for 173 two-party races for the US. House of Representatives in 1988. Let voteA be the
percentage of the votes received by candidate A, expendA and expendB the campaign expenditures
by candidate A and B respectively, and partystrA is a measure of party strength for Candidate A (the
percentage of the most recent presidential vote that went to A’s party). (a) Consider a model with an interaction between expenditures:
voteA = [30 + ﬁ1pﬂystrA + ﬁgexpendA + ﬁg expendB + ﬁ4expendA X expendB + at What is the partial effect of expendB on voteB, holding pnystrA and expendA ﬁxed? What is the
partial effect of expendA on voteA? Is the expected sign for [34 obvious? 1That is, going from 4 years to 5 years.
2Again, this is going from 19 to 20 years. ECO 321 2 of3 (b) Estimate the equation in part (a) and report the results in the usual form. Is the interaction
term statistically signiﬁcant? (c) Find the average of expend/l in the sample. Fix expend/i at 300 (for $300,000). What is the
estimated effect of another $100,000 spent by Candidate B on voteA? Is this a large effect? (d) Now ﬁx expendB at $100. What is the estimated effect of Aexpend = 100 on voteA? Does this
make sense? Estimate a 99% conﬁdence interval around the predicted change on vote/l. (e) Now, estimate a model that replaces the interaction with shareA, candidate A’s percentage share
of total campaign expenditures. Report the results in the usual form. Does it make sense to