Here is another one. Thanks all again for the help!
---------- Forwarded message ---------- From: Xiaogang Su <[email protected]> Date: Wed, Feb 26, 2014 at 9:38 AM Subject: Re: [R] Fitting glm with maxit=0? To: Thomas Lumley <[email protected]> Thanks much Prof. Lumley for the tip. That'll be great. Yeah, occasionally the (minus) Hessian matrix may not be even invertible. I will certainly check by first computing it with the standard formula and comparing. ============================= Xiaogang Su, Ph.D. Associate Professor Department of Mathematical Sciences University of Texas at El Paso 500 W. University Ave. El Paso, Texas 79968-0514 [email protected] [email protected] https://sites.google.com/site/xgsu00/ On Wed, Feb 26, 2014 at 9:22 AM, Thomas Lumley <[email protected]> wrote: > On Wed, Feb 26, 2014 at 2:24 AM, Xiaogang Su <[email protected]> wrote: > >> Dear All, >> >> Does anyone know if there is any way to obtain the variance-covariance >> matrix for any arbitrarily given estimates with the glm() function? >> > > Use maxit=1. The variance-covariance matrix is computed from the old > values, not the new values. > > You do need to check that the variance-covariance estimate is what you > want, since it's not designed for this use. Canonical link models with no > dispersion parameter should be ok, though. > > -thomas > > > -- > Thomas Lumley > Professor of Biostatistics > University of Auckland > [[alternative HTML version deleted]] ______________________________________________ [email protected] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
