On 30 Nov 2000 00:03:24 GMT, [EMAIL PROTECTED] (GJC) wrote: < snip, some details > > The applications known to me simply ignore this > specific deficiency of the observations. I wonder > whether it could be tractable? Of course, you > need some quantitative estimate of the error. > Weighted regression seems to me a correct way to > account for the fact that some observations are > less reliable than the others. Thus, the comment > by Rich is confusing for me. Am I wrong? Are > there more sophisticated ways to include such > error information in a predictive model? - I am glad the subject came up some more. I was pointing out that you cannot weight "Survive" differently from "Not-survive" and change any testing. But I am willing to add that you can do weighting for other purposes: to get fewer errors of one kind. Wholly arbitrary. Or you can do weighting based on other features: if one bird had been captured 10 times in the Pre period instead of once, perhaps it's absence in the Post is more notable. This problem of observing the birds does remind me of another problem with wildlife, the Capture-recapture paradigm. For here, I wonder if it might be modified to take into account a larger number of re-captures. I think my stats-FAQ has a reference or two on that problem in general > > > BTW: I am new in this NG. Is s.s.e. a good place > to discuss questions like the above, or the s.s.m., > or s.s.c. would be better? > For this question, I rank them - sci.stat.consult, sci.stat.edu, sci.stat.math. I put sci.stat.consult above s.s.e for this one, because the Q/A turns out to be a little messy for a classroom example. <That has to be a guess when you post, if you don't know the A. > -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html ================================================================= Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =================================================================
