similar to: model selection based on AICc

Hi, I'm fitting poisson regression models to counts of birds in 1x1 km squares using several environmental variables as predictors. I do this in a stepwise way, using the stepAIC function. However the resulting models appear to be overparametrized, since too much variables were included. I would like to know if there is the possibility of fitting models by steps but using the AICc

I would like test AICc as a criteria for model selection for a glm using stepAIC() from MASS package. Based on various information available in WEB, stepAIC() use extractAIC() to get the criteria used for model selection. I have created a new extractAIC() function (and extractAIC.glm() and extractAIC.lm() ones) that use a new parameter criteria that can be AIC, BIC or AICc. It works as

Hello, I'm trying to figure out how to run the stepAIC function starting with the NULL model. I can call the null model (e.g., lm(y ~ NULL)), but using this object in stepAIC doesn't seem to work. The objective is to calculate AICc. This can be done if stepAIC can be run starting with the NULL model; the (2p(p-1)/(n-p-1))to get AICc would be added to the final step AIC value. Can

Hi All: the package "MuMIn" can be used to select the model based on AIC or AICc. The code is as follows: data(Cement) lm1 <- lm(y ~ ., data = Cement) dd <- dredge(lm1,rank="AIC") print(dd) If I want to select the model by BIC, what code do I need to use? And when to select the best model based on AIC, what the differences between the function "dredge" in

Dear R-helpers I have generated a suite of GLMs. To select the best model for each set, I am using the meta-analysis approach of de Luna and Skouras (Scand J Statist 30:113-128). Simply put, I am calculating AIC, AICc, BIC, etc., and then using whichever criterion minimizes APE (Accumulated Prediction Error from cross-validations on all model sets) to select models. My problem arises where I

I am fitting logistic regression models, by defining my own link function, and would like to get AICc values. Using the glm command gives a value for AIC, but I haven't been able to get R to convert that to AICc. Is there a code that has already been written for this? Right now I am just putting the AIC values into an excel spreadsheet and calculating AICc, likelihood, and AIC

Hi, I am using 'best.arima' function from forecast package to obtain point forecast for a time series data set. The documentation says it utilizes AIC value to select best ARIMA model. But in my case the sample size very small - 26 observations (demand data). Is it the right to use AIC value for model selection in this case. Should I use AICc instead of AIC. If so how can I modify

Hi, I'm a very new user of R and I hope not to be too "basic" (I tried to find the answer to my questions by other ways but I was not able to). I have 12 response variables (species growth rates) and two environmental factors that I want to test to find out a possible relation. The sample size is quite small: (7<n<12, depending on each species-case). I performed a

I have a case-control study that I'm analysing using the conditional logistic regression function clogit from the survival package. I would like to calculate the AICc of the models I fit using clogit. I have a variety of scripts that can calculate AICc for models with a logLik method, but clogit does not appear to use this method. Is there a way I can calculate AICc from clogit in R? Many

How can I extract K (number of parameters) from an AIC calculation, both to report K itself and to calculate AICc? I'm aware of the conversion from AIC -> AICc, where AICc = AIC + 2K(K+1)/(n-K-1), but not sure of how K is calculated or how to extract that value from either an AIC or logLik calculation. This is probably more of a basic statistics question than an R question, but I thank

Dear R users, Would it be difficult to change the code of stepAIC (from the MASS library) to use AICc instead of AIC? It would be great to know of someone has tried this already. Best wishes Christoph.

Hi I get the following error when I try and get the AICc for a gls regression using qpcR: > AICc(gls1) Loading required package: nlme Error in n/(n - p - 1) : 'n' is missing My gls is like this: > gls1 Generalized least squares fit by REML Model: thercarnmax ~ therherbmax Data: NULL Log-restricted-likelihood: 2.328125 Coefficients: (Intercept) therherbmax 1.6441405

I am wondering if stepAIC in the MASS library may be used for model selection in an overdispersed Poisson situation. What I thought of doing was to get an estimate of the overdispersion parameter phi from fitting a model with all or most of the available predictors (we have a large number of observations so this should not be problematical) and then use stepAIC with scale = phi. Should this

Hi I'm trying to replicate Smith et al.'s (http://www.sciencemag.org/content/330/6008/1216.abstract) findings by fitting their Gompertz and logistic models to their data (given in their supplement). I'm doing this as I want to then apply the equations to my own data. Try as a might, I can't quite replicate them. Any thoughts why are much appreciated. I've tried contacting the

Folks, I'm using stepAIC(MASS) to do some automated, exploratory, model selection for binomial and Poisson glm models in R 1.3. Because I wanted to experiment with the small-sample correction AICc, I dug around in the code for the functions glm.fit stepAIC dropterm.glm addterm.glm extractAIC.glm and came across something I just don't understand. stepAIC() passes dropterm.glm() a

A few general comments about stepwiseAIC and a suggestion of how to select models a) Apart form the problem, that stepwise selection is not a garanty to get the best model, you need to have a lot of data to avoid overfitting if your model includes 7 parameter plus interactions (> 10 observations per parameter is what you are ideally looking for). b) Have a look at Anderson and Burnham's

Dear all, I have tried to modify the code of MASS:::stepAIC.default(), dropterm() and addterm() to use AICc instead of AIC for model selection. The code is appended below. Somehow the calculations are still not correct and I would be grateful if anyone could have a look at what might be wrong with this code... Here is a working example: ## require(nlme) model1=lme(distance ~ age + Sex, data =

Dear listers, Here we have a strange result we can hardly cope with. We want to compare a null mixed model with a mixed model with one independent variable. > lmmedt1<-lme(mediane~1, random=~1|site, na.action=na.omit, data=bdd2) > lmmedt9<-lme(mediane~log(0.0001+transat), random=~1|site, na.action=na.omit, data=bdd2) Using the Akaike Criterion and selMod of the package pgirmess

Hi Dr. Snow, I am a graduate student working on analyzing data for my thesis and came across your post on an R forum: The default link function for the glm poisson family is a log link, which means that it is fitting the model: log(mu) ~ b0 + b1 * x But the data that you generate is based on a linear link. Therefore your glm analysis does not match with how the data was generated

Hi all, I'm fitting GLM's using the step or stepAIC procedures and I would like to save the AIC of the intermediate models. I would appreciate very much information about how todo this. Best wishes Germ??n L??pez