5.6. Repeat Problem 5.4 or 5.5 for D 2, 5, 10, 20, and 50 multiple imputes, and compare answers. For what value of D

5.6. Repeat Problem 5.4 or 5.5 for D ¼ 2, 5, 10, 20, and 50 multiple imputes, and compare answers. For what value of D does the inference stabilize? PROBLEMS 91 5.7. Repeat Problem 5.4 or 5.5 using the more refined degrees of freedom formula (5.24) for the multiple imputation inference, and compare the resulting confidence intervals with the simpler intervals based on Eq. (5.23). 5.8. Apply the methods in Problems 5.1–5.5 to 500 replicate data sets generated as in Problem 5.2, and assess the bias of the estimates and the confidence coverage of intervals—valid intervals should exclude the true parameter in no more than about 50 of the 500 intervals. Interpret the results given your understanding of the properties of the various methods.