- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hello all,
I've recently been attempting to use SAS OnDemand to run a retrospective power analysis for longitudinal data initially modeled with SAS's proc mixed procedure to assess random and nested effects (Note that we are attempting a retrospective power analysis since we lacked pilot data to perform a proper prospective power analysis).
I am aware of some of the limitations associated with such analyses. In addition, I am aware that standard procedures in SAS, such as proc power or proc glmpower, are not appropriate for retrospective analyses. Also, I believe another issue that I am encountering is that I am not sure if the solutions that I have encountered will account for repeated or random effects (I have been interested in attempting the procedure described/macro here:https://support.sas.com/kb/25/011.html but the parameters derived from a glm may not be consistent with those derived from mixed models with random or repeated effects, if my understanding is correct) . Do any of you have any suggestions or recommendations? Thank you in advance!
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
I recommend getting a copy of SAS for Mixed Models. Either the 2nd edition or the newer Introduction and Basic Applications edition. Both have chapters on power and sample size determination (the newer version has two chapters on this topic). In any case, you feed in the variance components and an exemplary data set that builds in the differences you want to be able to detect, and run PROC MIXED holding random effects fixed with a PARMS statement. You then post-process the results to get an approximate noncentrality parameter. With that in hand, you can calculate the power. There are several worked examples.
SteveDenham