Re: Covariate Models Using Weight

Date: Mon, 22 Nov 1999 12:02:11 -0800 From: LSheiner <lewis@c255.ucsf.edu> Subject: Re: Covariate Models Using Weight Seems to me that this whole controversy is settled be acting like a Bayesian: the "pre-defined" scientifically based model has a certain prior credibility, which can be stated using an informative prior distribution on, for example, the allometric exponent. Then, if the data really favor some other value, the exponent will adjust; if they do not, it will remain near it's prior. Almost always, controversies like this can be resolved by choosing a "model" (in this case the full Bayesian framework) which is a superset of both of the special cases that find themselves in conflict ... LBS. -- Lewis B Sheiner, MD Professor: Lab. Med., Biopharm. Sci., Med. Box 0626 voice: 415 476 1965 UCSF, SF, CA fax: 415 476 2796 94143-0626 email: lewis@c255.ucsf.edu