M222 Modeling Business Decisions Fall 2017 1. You have collected data on 1559 movies released in the United States between 1995 and 2011. You have information on the following variables: The total gross revenue of the movie made as of May 22, 2013 in millions of Lifetime Gross 2011 dollars Lifetime Theaters The total number of theaters the movie is shown as of May 22, 2013 Year The year of Date of Release The estimated total cost of producing the movie, in millions of 2011 US Budget Dollars A numerical score assigned by Metacritic.com to the movie, ranging from 1 Metascore (worst reviews) to 10 (best reviews). You run run the following regression: (some numbers have been erased) Regression Statistics Multiple R 0.652432 R Square 0.425668 Adjusted R Square 0.42456 Standard Error 57.99512 Observations 1559 Coefficient s Intercept 1556.429 budget ($mil) 1.061 year -0.791 metascore 0.842 Standard Error t Stat 630.658 2.468 0.034 0.315 -2.513 0.082 10.220 P-value 0.014 Lower 95% 319.399 Upper 95% 2793.459 0.012 8.93E-24 -1.408 0.681 -0.174 1.004 a. Avatar, released in 2009, had a budget of 300 million dollars, a metascore of 8 and was released in December. Its gross revenue reached more than 2 billion dollars. Is Avatar as exceptionally successful as it sounds? Explain how you know. QM222 Modeling Business Decisions Fall 2017 b. Interpret the coefficient on Budget. Exactly what do we learn from it? c. What is the t-statistic associated with the coefficient on Budget? What does it tell us? d. Based on the coefficient on Budget, is investing in producing a movie profitable? Explain (based on the regression table). e. A corrupt employee of Metacritic.com offers to change the score of a movie from 5 to 8 and asks for 2.7 million dollars to do so. The (also corrupt) producer of the movie says that the amount is completely unreasonable. Do you agree or disagree with the producer's statement? Explain. f. There is an obvious reason why the coefficient on year is negative. What is this reason? QM222 Modeling Business Decisions Fall 2017 I have run a regression of lifetime gross on metascore along. That equation is (standard errors in parentheses): gross = 13.021 + 1.012 metascore (5.865) (0.105) g. Is the coefficient on metascore statistically significant? How do you know? h. Why is the coefficient on metascore so different in the two regressions? i. Which of these two regressions would be better to answer the question, \"How does having a better metascore affect the movie's gross lifetime revenues?\" -- the multiple regression or the simple regression - and why