Re: [R] a priori power analysis for glm, family = poisson

Probably the best way of doing this is via simulation. Have a look at chapter 5 of: Bolker, B. 2008. Ecological models and data in R. Princeton Univ. Press, Princeton, NJ. for an introduction and a bunch of examples. Joerg On Wed, Feb 20, 2013 at 5:49 AM, Chad Widmer wrote: > Dear R/statistic

Re: [R] Scatterplot and Causality

Nevertheless, some people try: http://www.kaggle.com/c/cause-effect-pairs Joerg On Mon, Apr 22, 2013 at 4:31 PM, Bert Gunter wrote: > Inline. > > > On Mon, Apr 22, 2013 at 1:15 PM, Lorenzo Isella > wrote: >> On Mon, 22 Apr 2013 17:27:05 +0200, Robert Baer wrote: >> >> >>> It strikes me that

Re: [R] Very small random effect estimation in lmer but not in stata xtmixed

I just realized that I have sent this only to Mohammed. So here it is to the list: These two model fits should yield the same results. In fact, if we use some simulated data generated by the model yik=2+0.5*xik1+0.25*xik2+uk+eik , with uk~N(0, 1) and eik~N(0, 1) and compare the results between S

Re: [R] Very small random effect estimation in lmer but not in stata xtmixed

cond(p<.5, 1, 0) gen eik=rnormal(0,3.5) gen yik=3.1+2.3*dum+uk+eik *-- On Thu, May 3, 2012 at 5:37 PM, Joerg Luedicke wrote: > I just realized that I have sent this only to Mohammed. So here it is > to the list: > > These two model fits should y