Question: You want to test H0 : 0 against H1 : 0. The test for H0 is to reject if Tn
You want to test H0 : µ Æ 0 against H1 : µ È 0. The test for H0 is to reject if Tn Æ bµ/s(bµ) È c where c is picked so that Type I error is ®. You do this as follows. Using the nonparametric bootstrap, you generate bootstrap samples, calculate the estimates bµ¤ on these samples and then calculate T ¤ Æ
bµ¤/s(bµ¤). Let q¤
1¡® denote the 1 ¡ ®th quantile of T ¤. You replace c with q¤
1¡®, and thus reject H0 if Tn Æ bµ/s(bµ) È q¤
1¡®. What is wrong with this procedure?
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
