D.4 Case 4

Predicting runs scored in baseball.

Consider a multiple regression model for predicting the total number of runs scored by a Major League Baseball (MLB) team during a season. Using data on number of walks (\(x_1\)), singles (\(x_2\)), doubles (\(x_3\)), triples (\(x_4\)), home runs (\(x_5\)), stolen bases (\(x_6\)), times caught stealing (\(x_7\)), strike outs (\(x_8\)), and ground outs (\(x_9\)) for each of the 30 teams during the 2014 MLB season, a first-order model for total number of runs scored (\(y\)) was fit.