Indermaur Lukas wrote: > Hi Frank > I fitted a set of 12 candidate models and evaluated the importance of > variables based on model averaged coefficients and SE (model weights >=0.9). > Variables in my models were not distributed in equal numbers across all > models thus I was not able to evaluate the importance of variables just by > summing up the AIC-weights of models including a specific variable. Now, why > so many models to fit: I was curious, if the ranking in the importance of > variables is similar, when just summing up the AIC-weights over an > all-possible-models set and looking at the ordered model averaged > coefficients (order of CV=SE/coefficient). > > Any hint for me? > Cheers > Lukas
I have seen the literature on Bayesian model averaging which uses weights from Bayes factors, related to BIC, but not the approach you are using. Frank > > > > Indermaur Lukas wrote: >> Hi, >> Fitting all possible models (GLM) with 10 predictors will result in loads of >> (2^10 - 1) models. I want to do that in order to get the importance of >> variables (having an unbalanced variable design) by summing the up the >> AIC-weights of models including the same variable, for every variable >> separately. It's time consuming and annoying to define all possible models >> by hand. >> >> Is there a command, or easy solution to let R define the set of all possible >> models itself? I defined models in the following way to process them with a >> batch job: >> >> # e.g. model 1 >> preference<- formula(Y~Lwd + N + Sex + YY) >> >> # e.g. model 2 >> preference_heterogeneity<- formula(Y~Ri + Lwd + N + Sex + YY) >> etc. >> etc. >> >> >> I appreciate any hint >> Cheers >> Lukas > > If you choose the model from amount 2^10 -1 having best AIC, that model > will be badly biased. Why look at so many? Pre-specification of > models, or fitting full models with penalization, or using data > reduction (masked to Y) may work better. > > Frank > >> >> >> >> >> °°° >> Lukas Indermaur, PhD student >> eawag / Swiss Federal Institute of Aquatic Science and Technology >> ECO - Department of Aquatic Ecology >> Überlandstrasse 133 >> CH-8600 Dübendorf >> Switzerland >> >> Phone: +41 (0) 71 220 38 25 >> Fax : +41 (0) 44 823 53 15 >> Email: [EMAIL PROTECTED] >> www.lukasindermaur.ch ______________________________________________ [email protected] 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.
