Generic question... I am familiar with generic power calculations in R, however a lot of the data I primarily work with is multivariate. Is there any package/function that you would recommend to conduct such power analysis? Any recommendations would be appreciated. Thank you for your time, Charles [[alternative HTML version deleted]]
On Mar 6, 2013, at 10:50 PM, Charles Determan Jr <deter088 at umn.edu> wrote:> Generic question... I am familiar with generic power calculations in R, > however a lot of the data I primarily work with is multivariate. Is there > any package/function that you would recommend to conduct such power > analysis? Any recommendations would be appreciated. > > Thank you for your time, > > CharlesAre you referring to a multivariate response or a multivariable model? Just trying to parse the terminology better. If the former, I don't believe that there is anything in R, but could be wrong. If correct, then you might want to look at simulation. If the latter, you might want to look at the Clinical Trials Task View: http://cran.r-project.org/web/views/ClinicalTrials.html as there are various packages that might fit what you need, but again, simulation is always an option. Regards, Marc Schwartz
On 3/6/2013 11:50 PM, Charles Determan Jr wrote:> Generic question... I am familiar with generic power calculations in R, > however a lot of the data I primarily work with is multivariate. Is there > any package/function that you would recommend to conduct such power > analysis? Any recommendations would be appreciated. > > Thank you for your time,I don't know an R implementation, but the theory of multivariate power analysis is well worked out and implemented in SAS/IML. See POWERLIB, described in http://www.jstatsoft.org/v30/i05 HTH -Michael -- Michael Friendly Email: friendly AT yorku DOT ca Professor, Psychology Dept. & Chair, Quantitative Methods York University Voice: 416 736-2100 x66249 Fax: 416 736-5814 4700 Keele Street Web: http://www.datavis.ca Toronto, ONT M3J 1P3 CANADA