Re: comparing theta's

From: LSheiner <lewis@c255.ucsf.edu> Date: Mon, 29 Mar 1999 11:05:29 -0800 Subject: Re: comparing theta's Which is simply to say that most frequentist hypothesis tests assume that the separate data points used to compute the test statistic (e.g., a mean) are INDEPENDENT. Postr Bayes estimates are not independent (each estimate depends on all others as each estimate is influenced by ALL the data), just as residuals are not independent (and hence simple usual statistics computed from them do not have the reference distribution that such statistics do have when they are computed from independent random variables). LBS. Lewis B Sheiner, MD Professor: Lab. Med., Biopharm. Sci., Med. Box 0626 UCSF, SF, CA 94143-0626 voice: 415 476 1965 fax: 415 476 2796 email: lewis@c255.ucsf.edu