(d) (5p) Use the regression coefficients that you computed in (c) to predict average test score for an out-of-sample candidate with Quality = 0.62 and GPA = 3.1. Verify that the prediction is identical to the prediction computed in (a). (e) (5p) Discuss the limitation of using OLS when k, the number of predictors, is large relative to (or even greater than) n when the prediction error, MSPE, is used to judge the quality of the model. There is a small approximation formula involving MSPE given in the Stock and Watson text that is likely to be helpful in your discussion. Then list the three alternative models to OLS presented by Stock and Watson that might produce better MSPE values and say a few words, perhaps including a formula, for each