I have a glm regression (quasi-poisson) of log(mu) on x but I have varying degrees of confidence in the x values, and can attach a numerical weighting to each. Can anyone help me with suggestions of how to analysise this. Is there an R package that would help? Wendy [[alternative HTML version deleted]]
Hi Wendy, In case you haven't see it, the glm function accepts an optional weights argument. Michael On 29 November 2010 09:42, Wendy Anderson <newhorizonscandelo at gmail.com> wrote:> I have a glm regression (quasi-poisson) of log(mu) on x but I have varying > degrees of confidence in the x values, and can attach a numerical weighting > to each. Can anyone help me with suggestions of how to analysise this. Is > there an R package that would help? > > Wendy > > ? ? ? ?[[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org 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. >
>> In case you haven't see it, the glm function accepts an optional >> weights argument. >> > > Thanks for the reply. But the philosopy behind weighting is the assumption > of unequal variance in the y values. In normal regression one assumes that > the x values are known without error > > WendySorry Wendy - I posted my reply prior to engaging my brain (ie. didn't read your question properly). You're talking about Model II / major axis type methods. The smatr package might cater for what you're trying to do. Hope this helps (more). Michael