Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Weichle, Thomas" <Thomas.Weichle@va.gov> |
To | <statalist@hsphsun2.harvard.edu> |
Subject | st: Folded F-Statistic |
Date | Wed, 8 Dec 2010 13:37:36 -0600 |
SAS, as Scott mentioned, provides the Folded F-Statistic in proc ttest. This information helps in deciding whether to use the ttest result based on the pooled variances or unpooled (unequal) variances (Satterthwaite method). The command sdtest appears to not follow the same convention as SAS when computing the folded F-Statistic. Stata uses group 0 variance in the numerator and group 1 variance in the denominator. As Scott mentioned, SAS's convention is to put the larger sample variance in the numerator and the smaller one in the denominator. Additionally, Stata uses the oneway command to perform the Bartlett's chi-squared test for equal variances. Both the Bartlett's test (oneway command) and Folded F-test (sdtest command) can be used but, as Steven mentioned, are sensitive to departures from normality. Thanks, Scott and Steven for your input. Tom Weichle Math Statistician Center for Management of Complex Chronic Care (CMC3) Hines VA Hospital, Bldg 1, C202 708-202-8387 ext. 24261 Thomas.Weichle@va.gov * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/