In: Statistics and Probability
Is a movie's earnings related to its ratings? Data on the top ten grossing movies in the United States for the weekend of July 13, 2007, were extracted from the Internet Movie Database (IMDB). These data included the weekend earnings and the average rating from users of IMDB. The rating system used by IMDB is a 10-star rating system. The following table shows the data. Using these data, determine whether there is a relation between a movie's earnings and its ratings.
Table: Top Ten Movies
|
There are n = 10 values for the variables. We shall use the t-test for the slope parameter for regression fit.
The regression output from excel is as below
Regression Statistics | ||||||||
Multiple R | 0.223348 | |||||||
R Square | 0.049884 | |||||||
Adjusted R Square | -0.06888 | |||||||
Standard Error | 24.46995 | |||||||
Observations | 10 | |||||||
ANOVA | ||||||||
df | SS | MS | F | Significance F | ||||
Regression | 1 | 251.5023 | 251.5023 | 0.420026 | 0.535078 | |||
Residual | 8 | 4790.228 | 598.7784 | |||||
Total | 9 | 5041.73 | ||||||
Coefficients | Standard Error | t Stat | P-value | Lower 95% | Upper 95% | Lower 95.0% | Upper 95.0% | |
Intercept | -19.3823 | 56.4848 | -0.34314 | 0.740335 | -149.636 | 110.8719 | -149.636 | 110.8719 |
X Variable 1 | 4.822115 | 7.440459 | 0.648094 | 0.535078 | -12.3356 | 21.97985 | -12.3356 | 21.97985 |
RESIDUAL OUTPUT | ||||||||
Observation | Predicted Y | Residuals | ||||||
1 | 18.7124 | 58.3876 | ||||||
2 | 19.19462 | 17.80538 | ||||||
3 | 22.08788 | -4.08788 | ||||||
4 | 19.19462 | -7.89462 | ||||||
5 | 5.692692 | 1.617308 | ||||||
6 | 17.26577 | -12.3358 | ||||||
7 | 9.550385 | -4.65038 | ||||||
8 | 19.67683 | -15.9968 | ||||||
9 | 21.60567 | -19.0057 | ||||||
10 | 15.81913 | -13.8391 |
Line of Best Fit Plot
Based on this regression line, the expected gross earnings for given movie ratings can be calculated as
Significance Test for Regression Fit
For the test of significance, the test statistic is
The critical value is obtained for degrees of freedom,
If we choose a significance level of 0.05, we have
Observed test statistic does not fall in the critical range, i.e. test statistic 0.6481 is less than the critical value viz. 2.306. Hence we conclude that there is no correlation between the two variables, and movie ratings cannot be used to predict its earnings.