On number 22, he went to the effort to determine the Rao-Cramer lower bound for the variance of the estimator. However, we already know the estimator is the sample mean and we know that the variance of the sample mean is the population variance divided by the sample size. The population variance is lambda^2 and so the variance of the estimator is lambda^2/n as he obtained.
In #29 he left out a word. The correct statement is "The posterior pdf will be proportional to the likelihood function times the prior pdf." He left out the words "times the prior pdf" but did do the multiplication in his solution.
In #16 he stated that the pdfs were discontinuous. That is not true. While it is true that for the uniform distribution the pdf is 0, then a constant, then 0 again, we do not call that discontinuous. When questions of continuity arise, it is only over the support, not over the entire line. However, because the two uniform distributions have *different* support, the mixture pdf is 0, then a constant, then a different constant, then 0 again. The lack of continuity arises from the change in constant.
------------------------------------------------------------------------------------------------
Stuart Klugman, FSA
Principal Financial Group Professor of Actuarial Science
Drake University
2507 University Avenue
Des Moines, IA 50311 USA
ph: 515-271-4097
e-mail: Stuart.Klugman@drake.edu