Rich Ulrich <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>... > > However, in particular, I don't remember this -- > Is there a particular ML estimation of "multiple > regression"? Is there an ML solution to the Normal > equations that isn't the OLS solution?
ML will generally differ from least squares when the assumed model for the random variable is something other than normal. Additionally, even if normality is assumed, there may still be a difference, since usually in regression the variance estimate is not ML (it usually has n-p rather than n as a divisor). Glen . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
