|
Activity Number:
|
283
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistical Education
|
| Abstract - #305292 |
|
Title:
|
Permutation Tests for Paired Samples: Power Comparison
|
|
Author(s):
|
Ferry Butar Butar*+ and Ananda Manage
|
|
Companies:
|
Sam Houston State University and Sam Houston State University
|
|
Address:
|
3657 red bud ln, Huntsville, TX, 77340,
|
|
Keywords:
|
Resampling ; wilcoxon ; permutation ; signed rank test
|
|
Abstract:
|
The t statistic is classic for testing a paired comparison when the distribution of the difference scores of random sample are normally distributed. For unspecified distributions, the sign test or the Wilcoxon signed rank test. Without knowledge of the underlying distribution, the permutation test can be utilized both for the original obsrvations as well as their ranks. For the permutation test, the significance level is exact when calculating all possible permutations. The approximate significance level is used when the numbers of permutations are very large. A simulation study is conducted to compare the power of the t, signed, and signed rank tests.
|
