econ problem (need some excel skills)

Problem Set 4

Due April 2 The following is cut and pasted from the syllabus: You are encouraged to work together on assignments, but must turn in your own assignment, in your own words. Verbatim copying is considered cheating. All assignments are expected to look professional. There are no late assignments or makeups accepted. Problem sets should be turned in as a hard copy in class. Hard copies handed in at class will receive up to 5 bonus points. (No scores over 100%) If you are unable to turn in a hard copy at class, you may turn in the assignment through Carmen. Problem sets turned in through Carmen are ineligible for bonus points and will receive limited feedback from the grader. The problem sets should look like professional reports. Therefore all problem sets are expected to meet the following criteria:

1. Everything is typed.

2. All tables and figures are named and numbered.

3. The necessary information to answer the question is concisely displayed in tables or figures when appropriate.

4. Tables and figures should be understandable without reading the text.

The first part of the problem set will be using the major league baseball attendance data on Carmen. The goal is to figure out if baseball is losing fans due to the length of games. Only use the data from 1946 to the present. Average game time is in an odd format due to the way excel read the table. You will need to create average game time in minutes. You can do this by highlighting the column, clicking text to columns under the data section, and formatting the cells to read as numbers. From there everyone should be able to construct minutes from hours and minutes. Report all regression results in clean, readable tables.

1. There has been a lot of discussion that baseball games are getting too long and baseball is losing fans because of it. Let’s test that theory. Estimate the equation: Attendance/game = β0 + β1AvgGameTime + e. Interpret the coefficient on minutes.

2. Its possible that games are running long because of more runs and fans like runs. Repeat the previous regression with runs per game included as an independent variable. Report the results and interpret both slope coefficients.

3. Another potential problem is the time trends in attendance and game time. It’s possible that attendance could be increasing only because population is increasing or for some other overall time trend. Create variables that are the difference in attendance and the difference in average game length.1 Interpret the regression coefficient. Is it statistically significant?

1For example in 2018 the average game was 180 minutes and in 2017 it was 184 minutes, therefore the difference in 2018 is -4. Be sure not to include the last year(1946).

4. Use the other variables to see if you can pinpoint causes of longer games. (There is no right or wrong answer here, but you have to run a regression that makes sense. Use 2 or more variables that would logically affect game time to run a regression and interpret the results.)

5. Next we will use the MLB team stats from the 2016, 2017, and 2018 seasons to assign values to players.Use the team stats file to find runs (column E) based on hits, doubles, triples, homeruns, stolen bases, caught stealing, walks, and strike- outs. Given that doubles, triples, and homeruns are controlled for, how is the coefficient on hits interpreted?

6. Next we will repeat the process for pitchers using runs against (Column S) as the dependent variable and hits, walks, homeruns, strikeouts, and hit by pitch as independent variables.

7. Use your regression results from number 5 to compare predicted runs produced by Miguel Cabrera in 2016 to predicted runs produced in 2018. Use regression results from number 6 to compare predicted runs allowed by Justin Verlander in 2016 and 2018.

8. Convert the differences to actual wins using the Pythagorean theorem from last homework. Suppose we have an average MLB team with 700 runs scored and 700 runs against currently winning 81 games( 700

2

7002+7002 ∗ 162 = 81). If that team

had 2018 Miguel Cabrera on it and you swapped out 2016 Miguel Cabrera, what would be the difference in wins? If each win is worth $9 million, then how much more is 2016 Cabrera worth to that team. Repeat with Verlander.