Multiplying this value by 2 if the observed test statistic was on the Sum(permutedTestStats <= observedTestStat)/N ThisĬomputes a “less than” p-value. A two-sided p-value could be obtainedīy multiplying this value by 2 if the observed test statistic was on the Value by 2 if the observed test statistic was on the left hand side ofĬomputes a “greater than” p-value. A two-sided p-value could be obtained by multiplying this ObservedTestStat)/N This computes a “less Test statistic was on the right hand side of the histogram. P-value could be obtained by multiplying this value by 2 if the observed myTest = observedTestStat)/N This computes a “greater than” p-value. Statistic needed to compute the p-value of the original test Test statistics then provides the sampling distribution of the test Test statistic on each permuted version of the data. (randomly rearranging) the data thousands of times and calculating a In review, the sampling distribution is created by permuting Test statistics that are as extreme or more extreme than the one
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |