1 / 4

APPENDIX B S OME B ASIC T ESTS IN S TATISTICS

Slides for Introduction to Stochastic Search and Optimization ( ISSO ) by J. C. Spall. APPENDIX B S OME B ASIC T ESTS IN S TATISTICS. Organization of appendix in ISSO Standard one-sample test P -values Confidence intervals Basic two-sample tests Matched pairs t -test

foy
Télécharger la présentation

APPENDIX B S OME B ASIC T ESTS IN S TATISTICS

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Slides for Introduction to Stochastic Search and Optimization (ISSO)by J. C. Spall APPENDIX BSOME BASIC TESTSIN STATISTICS Organization of appendix in ISSO Standard one-sample test P-values Confidence intervals Basic two-sample tests Matched pairs t-test Unmatched pairs t-test with identical variances Unmatched pairs t-test with nonidentical variances Other approaches to testing One- and two-sample tests important in stochastic search, optimization, and Monte Carlo simulation

  2. The Standard One-Sample Test • One set of data {Xi } for testing on   E(Xi) • Famous test statistics • z and t have a N(0, 1) and t-distribution, respectively • t-statistic useful in small samples; both z and t often used with non-normal samples • Large values of |z|or |t| indicate rejection of null hypothesis that  is some chosen value (commonly  = 0)

  3. P-Values • P-value:Probability that future experiment would have value of test statistic at least as extreme as that observed in the current experiment • Provides info. beyond binary accept/reject null hypothesis • Useful as indicator of strength of rejection • Example: If z = 2.15, P-value is 0.016 based on null hypothesis that   0 • Fairly strong evidence that  > 0

  4. Two-Sample Tests • Two sets of data {Xi } and {Yi } for testing X = Y • E.g.,Xi and Yi represent simulation outputs under two scenarios • Generic test statistic form where () denotes appropriate variance estimate • Three basic categories of tests affecting () • matched pairs • unmatched pairs; identical variances ( ) • unmatched pairs; non-identical variances ( ) • Large values of |t| indicate rejection of null hypothesis that X = Y

More Related