Pedro Martinez, who retired from Major League Baseball in 2012, had a stellar career, helping the Boston Red Sox to their first World Series title in 86 years in 2004. The next year he became a free agent and the New York Mets picked him up for $53 million for 4 years. Even after the move to New York, Martinez had his own fans. Possibly, he attracted more fans to the ballpark when he pitched at home, helping to justify his multimillion dollar contract. Was there really a €œPedro effect€ in attendance? We have data for the Mets home games of the 2005 season. The regression has the following predictors:
Weekend 1 if game is on a weekend day or night, 0 otherwise
Yankees 1 if game is against the Yankees (a hometown rivalry), 0 otherwise
Rain Delay 1 if the game was delayed by rain (which might have depressed attendance), 0 otherwise
Opening Day 1 for opening day, 0 for the others
Pedro Start 1 if Pedro was the starting pitcher, 0 otherwise Here€™s the regression.

a) All of these predictors are of a special kind. What are they called?
b) What is the interpretation of the coefficient for Pedro Start?
c) If we€™re primarily interested in Pedro€™s effect on attendance, why is it important to have the other variables in the model?
d) Could Pedro€™s agent claim, based on this regression, that his man attracts more fans to the ballpark? What statistics should he cite?

Transcribed Image Text:

Dependent variable is: Attendance R-squared-53.9% R-squared (adjusted)-50.8% s 6998 with 80 6 74 degrees of freedom Coeff ariable Intercept Weekend Yankees Rai Delay17427.9 7277 Opening Day 24766 7093 Pedro Start 5428.02 2017 SE(Coeff) t-ratio P-value <0.0001 6.15 <0.0001 0.0006 0.0192 0.0008 0.0088 28896.9 1161 9960.50 1620 15164.3 4218 24.9 3.59 -2.39 3.49 2.69