similar to: Lowest AIC after stepAIC can be lowered by manual reduction of variables

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

Hi list, I'm working with gamm models of this sort, using Simon Wood's mgcv library: gm<- gamm(Z~te(x,y),data=DATA,random=list(Group=~1)) gm1<-gamm(Z~te(x,y,by=Factor)+Factor,data=DATA,random=list(Group=~1)) with a dataset of about 70000 rows and 110 levels for Group in order to test whether tensor product smooths vary across factor levels. I was wondering if comparing those two

Hi all, I try to ask helps again about chan_h323 extensions. I define this in h323.conf: [general] port = 1720 bindaddr = 0.0.0.0 allow=gsm allow=ulaw gatekeeper = DISABLE context=default [gm1] type=friend host=192.168.1.20 context=default [gm2] type=friend host=192.168.1.25 context=default and I have in extensions.conf : [demo]

Dear All Sorry for what appears a trivial matter - I'm new to R and am stumbling ahead. I have a table of numerical data (36 rows by 12 columns) such as below: GM1 GM2 GM3 GM4 GM5 ...etc GM12 Run1 1 2 1 2 3 ... Run2 2 1 3 2 1 ... ... Run36 2 1 1 1 1 I would like to plot simple line graphs of some of the runs or

Hi all, I want to make extra columns in my datafile where the id of every groupmember is mentioned in separate columns. To explain it better see the example: id<-c(1,2,3,4,5,6,7,8,9,10,11,12) group<-c(1,1,1,1,2,2,3,3,3,3,3,3) a<-as.data.frame(cbind(id,group)) a id group 1 1 1 2 2 1 3 3 1 4 4 1 5 5

Dear R-helpers, this mail is intended to mention a rather trange result and generate potential useful comments on it. I am not aware of another posts on this issue ( RSiteSearch("quasipoisson lmer version dispersion")). MUsing the exemple in the reference of the lmer function (in lme4 library) and turning it into a quasi-poisson or quasi-binomial analysis, we get different results,

Hello all, Encouraged by the successful "demo", I'am getting on with Asterisk CVS. I added 2 H.323 extensions in extensions.conf [default] include => demo exten => 701,1,Dial(H323/gm2@192.168.1.20/s) exten => 702,1,Dial(H323/gm2@192.168.1.25/s) With: - [demo] is defined by default in sample.extensions.conf - Asterisk server is running on host 192.168.1.20, on the

Dear list, I am trying to use MuMIn to compare all possible mixed models using the dredge function on binomial data but I am getting an error message that I cannot decode. This error only occurs when I use glmer. When I use an lmer analysis on a different response variable every works great. Example using a simplified glmer model global model: mod<- glmer(cbind(st$X2.REP.LIVE,

Dear list, I am fitting a logistic multi-level regression model and need to test the difference between the ordinary logistic regression from a glm() fit and the mixed effects fit from glmer(), basically I want to do a likelihood ratio test between the two fits. The data are like this: My outcome is a (1,0) for health status, I have several (1,0) dummy variables RURAL, SMOKE, DRINK, EMPLOYED,

Dear list, I have another (again possibly boneheaded) puzzle about importing, again encapsulated in a nearly trivial package. (The package is posted at <http://www.math.mcmaster.ca/bolker/misc/coefsumtest_0.001.tar.gz>.) The package consists (only) of the following S3 method definitions: coeftab <- function(object, ...) UseMethod("coeftab",object) coeftab.default <-

Hi, I am working with gam models in the mgcv library. My response variable (Y) is binary (0/1), and my dataset contains repeated measures over 110 individuals (same number of 0/1 within a given individual: e.g. 345-zero and 345-one for individual A, 226-zero and 226-one for individual B, etc.). The variable Factor is separating the individuals in three groups according to mass (group 0,1,2),

I am currently attempting to hack the recently featured profileModels package so that it can handle models generated by the lme4 (mixed models) package. I'm getting really confused by different behavior of summary() before and after loading the lme4 package, and inside and outside the profileMethod() function. The basic behavior is that with lme4 loaded, and "obj" a fitted object

Hi, I'm working with gamm models of this sort: gm<- gamm(Z~te(x,y),data=DATA,random=list(Group=~1)) gm1<-gamm(Z~te(x,y,by=Factor)+Factor,data=DATA,random=list(Group=~1)) with a dataset of about 70000 rows and 110 levels for Group if I use plot(gm1$gam), I obtain 3 different surface plots, one for each level of my factor but I would like to create more complex contour plots for those 3

Hello, This file server was upgraded from 6.2-RELEASE-p$something to 6.3-RELEASE-p3 (current RELENG_6_3 sources via csup) on August, 11 and rebooted to the new kernel on August, 12. The school is closed due to holidays. The only load is mail delivery to Maildirs via NFS from another server (just 50 mailboxes or so, none of them very busy) and nightly backups. I went there on August, 13 to perform

I am currently running a generalized linear mixed effect model using glmer and I want to estimate how much of the variance is explained by my random factor. summary(glmer(cbind(female,male)~date+(1|dam),family=binomial,data= liz3")) Generalized linear mixed model fit by the Laplace approximation Formula: cbind(female, male) ~ date + (1 | dam) Data: liz3 AIC BIC logLik deviance 241.3

Dear all, I found a problem with 'lme4'. Basically, once you load the package 'aod' (Analysis of Overdispersed Data), the functions 'lmer' and 'glmer' don't work anymore: library(lme4) (fm1 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy)) (gm1 <- glmer(cbind(incidence, size - incidence) ~ period + (1 | herd), family = binomial, data

https://bugzilla.netfilter.org/show_bug.cgi?id=1740 Bug ID: 1740 Summary: hashlimit limit: reduction to lowest terms in the output is confusing Product: iptables Version: 1.8.x Hardware: x86_64 OS: Debian GNU/Linux Status: NEW Severity: minor Priority: P5 Component:

Hello, everyone I have a piece of code that looks like this: mrets <- merge(mrets, BMM.SR=apply(mrets, 1, MyFunc, ret="BMM.AV120", stdev="BMM.SD120")) mrets <- merge(mrets, GM1.SR=apply(mrets, 1, MyFunc, ret="GM1.AV120", stdev="GM1.SD120")) mrets <- merge(mrets, IYC.SR=apply(mrets, 1, MyFunc, ret="IYC.AV120",

Based on the stepAIC help, I have assumed that it only was for lm, aov, and glm models. I gather from the following correspondence that it also works with lme models. Thomas Lumley 07:40 a.m. 28/04/03 -0700 4 Re: [R] stepAIC/lme problem (1.7.0 only) Prof Brian Ripley 04:19 p.m. 28/04/03 +0100 6 Re: [R] stepAIC/lme problem (1.7.0 only) Prof Brian Ripley 06:09 p.m. 29/04/03 +0100 6 Re: [R]

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