Consider the following output from a Cumulative Logit model where X is the person's age and Y: 1 - if the person went to the movies less than 5 times during the year; 2 - went between 5 and 10 times; 3 - went more than 10 times (4 marks each):

(a) Why was the Cumulative Logit used instead of the Multinomial Logit?

(b) Based on the output, was the correct model used?

(c) Why are there 2 intercept estimates and only 1 slope estimate?

(d) Are any of the coefficients significant?

(e) Provide an interpretation of the coefficients that are significant

Score Test for the Proportional Odds Assumption Chi-Square DF Pr > ChiSq 1.2813 0.2577 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Exp(Est) Intercept 3 1 1.1597 0.7713 2.2607 0.1327 3.189 Intercept 2 1 3.1191 0.8290 14.1559 0.0002 22.625 X 1 -0.0916 0.0257 12.6987 0.0004 0.912