Economic Analysis: Predicting U.S. Box Office Revenue Determinants

This report analyzes the predictors of U.S. domestic box office revenue using regression analysis. The study focuses on the top 30 highest-grossing English films in the U.S. and examines the impact of positive and negative reviews, and release dates. The findings indicate that these variables are statistically insignificant in predicting box office revenue, as evidenced by high p-values. The report also discusses the model, hypothesis, and data sources (Rotten Tomatoes). The conclusion suggests that the null hypothesis should be failed to be rejected, and recommends further research considering other potential predictor variables such as film ratings. The assignment also outlines tasks for team members to enhance the introduction, closing, data model, and correlation coefficients.

Analysis of U.S. Box Office Revenue Predictors
Abstract
This paper examines the effect of various predictor variables domestic box office
revenue. The sample space in this study is comprised of the top 30 highest-grossing English
films produced in the United States of all time. Regression results indicate that the amount of
positive and negative reviews, along with the release date are all statistically insignificant based
on their low p-values from the regression analysis output.
Introduction
The rising costs of movie productions have resulted in motion picture studios seeking to
understand the determinants of a successful and large revenue generating movie. The purpose of
this research is to analyze the film industry with a concentration on the determinants of domestic
box office revenue for English language movies. The importance of the film industry cannot be
ignored based on the twenty-nine billion dollars in industry revenue in 2015.
Data and Model
The primary source of data for this study is the Rotten Tomatoes website. Rotten
Tomatoes has a unique rating system that summarizes positive or negative reviews of accredited
film critics for each motion picture. In addition to providing a system of aggregate reviews,
Rotten Tomatoes also contains information pertaining to box office revenue, release date, and
film ratings. Our model for this study is specified as:
Where Y is the domestic box office earnings, ………………...

Our hypothesis for this study is specified as:
Null Hypothesis: H0: β1 = β2 = … = βp-1 =0. The initial assumption is that there is no relation
between the amount of positive and negative movie reviews along with the time of release to the
domestic box office revenues.
Alternative Hypothesis: H1: At least one βi is ≠ 0. At least one of the independent variables is
useful in explaining or predicting domestic box office revenues.
Predictors of Domestic Box Office Revenue
The variables “Positive Reviews” and “Negative Reviews” are the number of approval or
disapproval ratings for a film by a leading group of movie reviewers. Conventional wisdom
suggests that critical reviews are extremely important to the popularity of movies, especially in
the early stages of a release. Positive reviews are expected to attract consumers and identify
quality, while negative reviews are expected to limit the interest of the influential early adopters.
The multiple regression analysis revealed that neither the amount of positive or negative reviews
were statistically significant in predicting a film’s box office revenue.
Conclusion
The most important result of the study is the observation that the amount of positive and
negative reviews, along with the release date were all statistically insignificance based on their
high p-values from the regression analysis. The p-values of the predictor variables, low R2 value,

and low F statistic are an indicators that we should fail to reject the null hypothesis. All of the
independent variable revealed that their was no relation between the amount of positive and
negative movie reviews along with the time of release to the domestic box office revenues.
Further research into the cause of a film’s financial performance could be conducted by
testing additional predictor variables. A movie’s film rating could potentially affect the financial
performance of a film. The rating is assigned by the Motion Picture Association of America,
which establishes a rating code as a means of giving advance information to parents and others
about the theme and treatment of films. Conventional wisdom leads an audience to believe that
family friendly products sell, while an adult theme has a limited customer base because of age
restrictions preventing access to the lucrative teenage market. Further research could be
conducted to determine if any additional predictor values are statistically significant in
contributing to a movies box office success.
TO DO:
Anna:
Intro and Closing to one page each
expand on interpretation of the results
prezi
why the model is significant to the industry
Saranya:
Prezi
Expand on determinates
Calculation of the residual values
discuss any outliers

discuss no regression
Alikhan:
Inserting and properly labeling/explaining
Expand on data model
Correlation coefficient
- Have done tomorrow at 7 AM
- → Email to Prof Aguirre before 8 AM
- Practice presentation at 9 PM on Tuesday