For various reasons, I spent part of my time today looking at sample
size and power calculation tools (don't ask, don't tell...). This
seems to be one area that R is incredibly weak in (well, nearly all
stat packages, except perhaps specialized tools and SAS); sure, there
are a number of functions in various packages:
base, statmod, Hmisc
Have I missed something? (I would've expected at least one sequential
computation, or non-standard design, but apparently there are none, or
I missed it).
I'd appreciate hearing about work that I've missed...
best,
-tony
--
rossini at u.washington.edu http://www.analytics.washington.edu/
Biomedical and Health Informatics University of Washington
Biostatistics, SCHARP/HVTN Fred Hutchinson Cancer Research Center
UW (Tu/Th/F): 206-616-7630 FAX=206-543-3461 | Voicemail is unreliable
FHCRC (M/W): 206-667-7025 FAX=206-667-4812 | use Email
CONFIDENTIALITY NOTICE: This e-mail message and any attachme...{{dropped}}
rossini at blindglobe.net (A.J. Rossini) writes:> For various reasons, I spent part of my time today looking at sample > size and power calculation tools (don't ask, don't tell...). This > seems to be one area that R is incredibly weak in (well, nearly all > stat packages, except perhaps specialized tools and SAS); sure, there > are a number of functions in various packages: > > base, statmod, Hmisc > > Have I missed something? (I would've expected at least one sequential > computation, or non-standard design, but apparently there are none, or > I missed it). > > I'd appreciate hearing about work that I've missed...Is SAS particularly hot? I've just been explaining to people how to cheat SAS Analyst into letting the Two-Sample t-Test sample-sizer do binomial proportions with a fudged SD. I think Claus Ekstr?m submitted a modification power.t.test for unequal sample size, but he wasn't sufficiently assertive that it was actually done right, so it didn't get in. One thing we really ought to get around to soon is sample sizing for equivalence studies. And, yes, sequential procedures would be obvious too. -- O__ ---- Peter Dalgaard Blegdamsvej 3 c/ /'_ --- Dept. of Biostatistics 2200 Cph. N (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk) FAX: (+45) 35327907
Dear Peter, At 12:30 PM 11/11/2003 +0100, Peter Dalgaard wrote:>rossini at blindglobe.net (A.J. Rossini) writes: > > > For various reasons, I spent part of my time today looking at sample > > size and power calculation tools (don't ask, don't tell...). This > > seems to be one area that R is incredibly weak in (well, nearly all > > stat packages, except perhaps specialized tools and SAS); sure, there > > are a number of functions in various packages: > > > > base, statmod, Hmisc > > > > Have I missed something? (I would've expected at least one sequential > > computation, or non-standard design, but apparently there are none, or > > I missed it). > > > > I'd appreciate hearing about work that I've missed... > >Is SAS particularly hot? I've just been explaining to people how to >cheat SAS Analyst into letting the Two-Sample t-Test sample-sizer do >binomial proportions with a fudged SD.There is an extensive set of power-calculation macros in SAS, by Ralph O'Brien, at <http://www.bio.ri.ccf.org/power.html>. Apparently SAS Version 9.1 will have new procs power and glmpower, which will cover most of what is in these macros. Regards, John ----------------------------------------------------- John Fox Department of Sociology McMaster University Hamilton, Ontario, Canada L8S 4M4 email: jfox at mcmaster.ca phone: 905-525-9140x23604 web: www.socsci.mcmaster.ca/jfox