similar to: need help getting started writing a new varFunc class for lme()

Hello - I'm trying to modify an option for the lme() or nlme() macros. I want to write my own specification for the variance function and am following homework problem 4, Chapter 5, page 268 of Pinheiro and Bates book on mixed effect. I'm up to point where I've created a new class using an existing variance function class, varExp as a template. Next I need to write an

Dear R-help members, I have a question regarding how to use varComb function to specify a variance function for the "weights" in the gls. I need to fit a linear model with heteroscedasticity. The variance function is exp(c0+nu0*W +nu1*W^2) where W is a covariate. Initially I want to use varFunc to define my own variance function following the instruction in the Pinheiro and

Dear R-help, I am trying to set up linear mixed effects models in R using the (recommended) nlme package (R version 2.3.1 on a Linux platform). When trying to reproduce an example from Jose Pinheiro & Douglas Bates (2000, p 210) I get the following error message (code to produce message pasted as well): library("nlme") data("Orthodont") vf1Ident <- varIdent(

Hello. I am trying my hand at modifying the varFunc class varExp, but I must be missing a step. All I want to do right now is make a working copy of varExp, call it varExp2, and then later change it. coef.varExp2, coef<-.varExp2, and Initialize.varExp2 all seem to work properly after I construct them. I can successfully use the commands: v2 <- varExp2(form = ~age|Sex,fixed =

Dear R users, I am currently analyzing a dataset using lme(). The model I use has the following structure: model<-lme(response~Covariate+TreatmentA+TreatmentB,random=~1|Block/Plot,method="ML") When I plot the residuals against the fitted values, I see a clear positive trend (meaning that the variance increases with the mean). I tried to solve this issue using weights=varPower(),

Hi, I am running R2.8.1 under Linux, and I am having trouble using the variance functions in nlme My basic model was something like: model0 <- lme( log(growth) ~ light * species.group , data=data, random=~light|species ) # with 20 odd species divided in 2 groups Following the methods in Pinheiro&Bates I tried to put a variance function in the model: model1 <- update(model0,

Dear R-help members, I have a question regarding how to use varComb function to specify a variance function for the "weights" in the gls. I need to fit a linear model with heteroscedasticity. The variance function is exp(c0+nu0*W +nu1*W^2) where W is a covariate. Initially I want to use varFunc to define my own variance function following the instruction in the Pinheiro and Bates

Hi, I would like to fit a model for a factorial design that allows for unequal variances in all groups. If I am not mistaken, this can be done in lm by specifying weights. A function intended to specify weights for unequal variance structures is provided in the nlme library with the varIdent function. Is it apropriate to use these weights with lm? If not, is there another possibility to do

Greetings R-helpers, I am attempting to fit an lme() while specifying a correlation structure, but I'm getting into trouble long before I get to that point. I am receiving the error: Error in y[revOrder] - Fitted : non-conformable arrays It doesn't seem to matter how simple or complex the model I specify is, it always gives this same error message. This makes me suspect something is

Hello all. I was wondering if there is any way to adjust the denominator degrees of freedom in lme(). It seems to me that there is only one method that can be used. As has been pointed out previously on the list, the denominator degrees of freedom given by lme() do not match those given by SAS Proc Mixed or HLM5. Proc Mixed, for example, offers five different options for computing the

HI Dave, My comment was based? on: " >The main question with this test was if the interaction term is significant (i.e. growth rate). However, my question is could I also look at the p-values of the main effects to ?>say if body mass increase significant with body mass?" Here, the result shown were from the summary of the linear model.?? We report the p-values of the main

I am using rw0641. In my continuing quest to understand repeated measures analysis, I again return to lme. I exported the Potthoff and Roy data Orthodont.dat from S-PLUS 4.5 to avoid capture errors and ran the examples in the R help. I imported the data.frame with data <- read.table("Orthodont.dat",header=T) attach(data) and created the objects Orthodont.fit1 <-

Hi, we have a split-plot experiment in which we measured the yield of crop fields. The factors we studied were: B : 3 blocks I : 2 main plots for presence of Irrigation V : 2 plots for Varieties N : 3 levels of Nitrogen Each block contains two plots (irrigated or not) . Each plot is divided into two secondary parcels for the two varieties. Each of these parcels is divided into three subplots

Hello- I understand that it's convention, when comparing two models using the anova function anova(model1, model2), to put the more "complicated" (for want of a better word) model as the second model. However, I'm using lme in the nlme package and I've found that the order of the models actually gives opposite results. I'm not sure if this is supposed to be the case

Dear list, I was trying to create a VarClass for nlme to work with Fay-Herriot (FH) models. The idea was to create a modification of VarComb that instead of multiplying the variance functions made their sum (I called it varSum). After some fails etc... I found that the I was not getting the expected results because I needed to make sigma fixed. Trying to find how to make sigma fixed I run into

There has been some recent discussion on this list about the value of using intervals with lme() to check for whether a model is ill-defined. My question is, what else can drive very large confidence intervals for the variance components (or cause the error message "Error in intervals.lme(Object) : Cannot get confidence intervals on var-cov components: Non-positive definite approximate

Dear R-users, I have longitudinal data and would like to fit a model where both the variance-covariance matrix of the random effects and the residual variance are conditional on a (binary) grouping variable. I guess the model would have the following form (in hierarchical notation) Yi|bi,k ~ N(XiB+Zibi, sigmak*Ident) bi|k ~ N(0, Dk) K~Bernoulli(p) I can obtain different sigmas (sigma0 and

This question comes up periodically, probably enough to give it a proper thread and maybe point to this thread for reference (similar to the 'conservative anova' thread not too long ago). Moving from lme syntax, which is the function found in the nlme package, to lmer syntax (found in lme4) is not too difficult. It is probably useful to first explain what the differences are between the

You wrote: Ø Dear All, Ø Can anyone provide me with reference notes(or steps) towards analysis of?? (un)balanced panel data in R. Ø Thank you! The "plm" package does panel data analysis in R. See the vignette at: cran.r-project.org/web/packages/plm/vignettes/plm.pdf. There are other similar articles by the same authors, Yves Croissant and Giovanni Millo, and one of these is the

I have a longitudinal data analysis project. There are 10 observations on each of 15 units, and I'm estimating this with randomly varying intercepts along with an AR1 correction for the error terms within units. There is no correlation across units. Blundering around in R for a long time, I found that for linear/gaussian models, I can use either the MASS method glmmPQL (thanks to