similar to: Error with corCompSymm and lme fit for repeated measures

Dear R-devel members, While writing a new correlation structure similar to 'corCompSymm' and intended to be used with 'gnls', I got puzzled with the 'Initialize' method. Using 'Initialize' before 'gnls' may be regarded as a mean to set an initial value for the corStruct parameter. However 'gnls' does not work properly with a

We are trying to use the corCompSymm function in nlme The example from the help pages for the corAR1 function gives the following > corAR1(0.2, form = ~ 1 | Mare) Correlation structure of class corAR1 representing Phi 0.2 We are expecting a somewhat similar correlation specification with the help page example for corCompSymm, but just get an error instead > corCompSymm(0.5, form = ~ 1 |

Hi, I have confirmed temporal correlation problems in my data. Is there a possibility to use corCompSymm for a gamm()? I am an R-beginner. I have very short time series. There are three years and within each year, there are 10 weeks. he 10 weeks are the same every year and have not unique values, I seem not to be able to use AR-1 (I assume that I have too little data for autoregression

Hello, today i typed "htop" for controlling my ressources. I could see that my swap is neraly 100%. This problem occurs since start of the server, about 3 year ago. Its not a critical issue for me, because the server is running fine. Several times i incereased the size of swap. Today 9,3GB of 10GB swap are allocated (33 day uptime). My system is still running and i have no

Dear All, I am analysing a dataset on levels of herbivory in seedlings in an experimental setup in a rainforest. I have seven classes/categories of seedling damage/herbivory that I want to analyse, modelling each separately. There are twenty maternal trees, with eight groups of seedlings around each. Each tree has a TreeID, which I use as the random effect (blocking factor). There are two

Dear list, I need to include in a superscript number (-1) in a y-axis label of a boxplot graphic. I tried the following line: >boxplot(five$gr13~five$Code, xlab="Species code", ylab=substitute(paste("DGL growth (cm.yr"^{-1},")"))) It returns the following error message: Error in "DGL growth (cm.yr"^{ : non-numeric argument to binary operator I used

Hi all I am trying to generate a normal unbalanced data to estimate the coefficients of LM, LMM, GLM, and GLMM and their standard errors. Also, I am trying to estimate the variance components and their standard errors. Further, I am trying to use the likelihood ratio test to test H0: sigma^2_b = 0 (random effects variance component), and the t-test to test H0:mu=0 (intercept of the model Yij = mu

Hello, R-users,   I have a datafile with 37313 records and each record has 5 different measurements on the same variables. The format looks like this: treeID, VIG0, VIG1, VIG2, VIG3, VIG4 I was trying to convert the one row record to 5 rows record with format like this (treeID, MEASUREMENT, VIGOR). My code like this:   treeMeas<-matrix(data=0,nrow=(length(tree1$indivTree)*5), ncol=3)

Dear list members, I'm interested in showing that within-group statistical dependence is negligible, so I can use ordinary linear models without including random effects. However, I can find no mention of testing a model with vs. without random effects in either Venable & Ripley (2002) or Pinheiro and Bates (2000). Our in-house statisticians are not familiar with this, either,

Dear R-help members, I'm trying to run LME model on some behavioral data and need confirmations about what I'm doing... Here's the story... I have some behavioral reaction time (RT) data (participants have to detect dome kind of auditory stimuli). the dependant variable is RT measured in milliseconds. 61 participants were tested separated in 4 age groups (unblanced groups,

Hi everyone, I'm using lmer and estimable (from packages lme4 and gmodels respectively) and have the disconcerting happening that when I run exactly the same code, I get different results! In checking this out by running the code 50x, it seems to be that answers may be randomly deviating around those which I get from another stats package (GenStat, using the linear mixed models functionality

Hi folks, I am having trouble to plot a mixed model analysis of covariance (ANCOVA). To do so I use the function predict in nlme but the line that is being drawn is totally out of control!!! here is my script (where MASS_S is dry mass and MASS_F is fresh mass): MEN<-read.table("Mentha_lme2.txt", h=T) attach(MEN) lme1<-lme(log(MASS_S)~log(MASS_F)*TREAT, random=~1|INDIV)

Dear Friends, I am seeking for any help on an error message in lme functions. I use mixed model to analyze a data with compound symmetric correlation structure. But I get an error message: "Error in corMatrix.corCompSymm(object) : NA/NaN/Inf in foreign function call (arg 1)". If I change the correlation structure to corAR1, then no error. I have no clue how to solve this problem.

Dear R-helpers, When I try > anova(txtE2.lme, txtE2.lme1) Model df AIC BIC logLik Test L.Ratio p-value txtE2.lme 1 10 8590 8638 -4285 txtE2.lme1 2 7 8591 8624 -4288 1 vs 2 6.79 0.0789 > latex(anova(txtE2.lme, txtE2.lme1)) Error: object "n.group" not found I don't even see n.group as one of the arguments of latex() I checked to see >

Hi, I am having trouble using the lme function to perform a nested ANOVA with a random nested factor. My design is as follows: Location (n=6) (Random) Site nested within each Location (n=12) (2 Sites nested within each Location) (Random) Dependent variable: sp (species abundance) By using the aov function I can generate a nested ANOVA, however this assumes that my nested

R2.2 for WinXP, Splus 6.2.1 for HP 9000 Series, HP-UX 11.0. I am trying to get a handle on why the same lme( ) code gives such different answers. My output makes me wonder if the fact that the UNIX box is 64 bits is the reason. The estimated random effects are identical, but the fixed effects are very different. Here is my R code and output, with some columns and rows deleted for space

Dear Thomas Lumley, and R-help list members, I have read your article "Network meta-analysis for indirect treatment comparisons" (Statist Med, 2002) with great interest. I found it very helpful that you included the R code to replicate your analysis; however, I have had a problem replicating your example and wondered if you are able to give me a hint. When I use the code from the

Dear all, Is it possible to incorporate a nested design in GEE? I have measurements on trees that where measured in two years. The trees are nested in plots. Each plot contains 24 trees. The number of plots is 72. Hence we would expect 2 * 24 * 72 = 3456 data points. A few are missing, so we end up wih 3431 data points. This is what I have tried until now. #assuming independence between trees

Dear members, I would like to switch from nlme to lme4 and try to translate some of my models that worked fine with lme. I have problems with the pdMat classes. Below a toy dataset with a fixed effect F and a random effect R. I gave also 2 similar lme models. The one containing pdLogChol (lme1) is easy to translate (as it is an explicit notation of the default model) The more parsimonious

Hi, I have searched the archives and can't quite confirm the answer to this. I appreciate your time... I have 4 treatments (fixed) and I would like to know if there is a significant difference in metal volume (metal) between the treatments. The experiment has 5 blocks (random) in each treatment and no block is repeated across treatments. Within each plot there are varying numbers of