Question 1

Which one of the following statements is FALSE?

(a) Two common purposes for fitting regression models are for explanation or prediction.

(b) Cross-validation can be used to get a realistic estimate of prediction error when the data set is not large enough to be divided into a training set and a test set.

(c) Using the same data to fit a predictive model and to assess the prediction error rate tends to result in an estimate that is too high.

(d) The "dredge" function can be used to fit all possible subset models using BIC as the model selection criterion.

(e) For large sample sizes, AIC and AICc will give almost the same values.

Question 2

Suppose that W has a chi-squared distribution with 24 degrees of freedom.

We wish to estimate the probability Pr(W > 26) using simulation. What R command will achieve this? Assume that N is a verylarge number.

(a) mean(rchisq(N, 26) > 24)

(b) var(rchisq(N, 26) > 24)

(c) mean(rchisq(N, 26) < 24)

(d) var(rchisq(N, 24) > 26)

(e) mean(rchisq(N, 24) > 26)

(f) mean(rchisq(N, 24) < 26)