similar to: discrepancy between paired t test and glht on lme models

Hello! I have larger and a smaller data frame with 1 factor in each - it's the same factor: large.frame<-data.frame(myfactor=LETTERS[1:10]) small.frame<-data.frame(myfactor=LETTERS[c(9,7,5,3,1)]) levels(large.frame$myfactor) levels(small.frame$myfactor) table(large.frame$myfactor) table(small.frame$myfactor) myfactor has 10 levels in large.frame and 5 levels in small.frame. All 5

Hi, Does tapply change the order when applied on a factor? Below is the code I tried. > mylevels<-c("IN0020020155","IN0019800021","IN0020020064") >

I have a data frame with some factor columns. I want to drop the rows with rare factor values (and remove the factor values from the factors). E.g., frame$MyFactor takes values A 1,000 times, B 2,000 times, C 30 times and D 4 times. I want to remove all rows which assume rare values (<1%), i.e., C and D. i.e., frame <- frame[[! (frame$MyFactor %in% c("A","B"))]] except

Dear R help group, I am teaching myself linear mixed models with missing data since I would like to analyze a stats design with these kind of models. The textbook example is for the procedure "proc MIXED" in SAS, but I would like to know if there is an equivalent in R. This example only includes two time-measurements across subjects (a t-test "with missing values"), but I

Hello, I have a list of data with multiple elements, and each element in the list has multiple variables in it. Here's an example: ### Make the fake data dv <- c(1,3,4,2,2,3,2,5,6,3,4,4,3,5,6) subject <- factor(c("s1","s1","s1","s2","s2","s2","s3","s3","s3",

Hi, I searched for this in the mailing list, but found no results. I have a large dataframe ( dim(mydata)= 1297059 16, object.size(mydata= 145280576) ) , and I want to perform some calculations which can be done by a factor's levels, say, mydata$myfactor. So what I want is to split this dataframe into nlevels(mydata$myfactor) = 80 levels. But I must do this efficiently, that is, I

Dear All, I would like to run a generalized linear mixed model with the software R (one categorical predictor, one random factor, the distribution of the dependent variable is binomial, and the link is logit). Thereafter, I would like to perform multiple comparisons (post hoc test) among the groups of the categorical predictor. Is it possible with the software R? Are traditional methods of

Hi, I want to export the result of glht in R into a LaTeX table, such as that result: Linear Hypotheses: Estimate Std. Error z value Pr(>|z|) Group1 - Group2 == 0 -0.14007 0.01589 -8.813 <0.001 "***" Group1 - Group3 == 0 -0.09396 0.01575 -5.965 <0.001 *** --- Signif. codes: 0 ?***? 0.001 ?**? 0.01 ?*? 0.05

Hi, I am experimenting with using glht() from multcomp package together with coxph(), and glad to find that glht() can work on coph object, for example: > (fit<-coxph(Surv(stop, status>0)~treatment,bladder1)) coxph(formula = Surv(stop, status > 0) ~ treatment, data = bladder1) coef exp(coef) se(coef) z p treatmentpyridoxine -0.063 0.939 0.161

Hello, I am trying to determine the most appropriate way to run post-hoc comparisons on my lme model. I had originally planned to use Tukey HSD method as I am interested in all possible comparisons between my treatment levels. TukeyHSD, however, does not work with lme. The only other code that I was able to find, and which also seems to be widely used, is glht specified with Tukey:

Dear everybody, I have the following challenge. I have a data set with 2 subgroups, dates (days), and corresponding values (see example code below). Within each subgroup: I need to aggregate (sum) the values by week - for weeks that start on a Monday (for example, 2008-12-29 was a Monday). I find it difficult because I have missing dates in my data - so that sometimes I don't even have the

Hi all: There's a question about glht function. My data:data_ori,which inclue CD4, GROUP,time. f_GROUP<-factor(data_ori$GROUP) f_GROUP is a factor of 3 levels(0,1,2,3) result <- lme(sqrt(CD4) ~ f_GROUP*time ,random = ~time|ID,data=data_ori) glht(result, linfct = mcp(f_GROUP="Tukey") ) Error in `[.data.frame`(mf, nhypo[checknm]) : undefined columns selected I can't

Hi, I am trying to test some contrasts, using glht() in multcomp package on fixed effects in a linear mixed model fitted with lme() in nlme package. The command I used is: ## a simple randomized block design, ## type is fixed effect ## batch is random effect ## model with interaction dat.lme<-lme(info.index~type, random=~1|batch/type, data=dat) glht(dat.lme, linfct = mcp(type

Using a factor with 'number' levels the straightforward symbolic formulation of a contrast in 'glht' of the 'multcomp' package fails. How can this problem be resolved without having to redefine the factor levels? Example: #A is a factor with 'number' levels #B similar factor with 'letter' levels dat<-data.frame(y=1:4,A=factor(c(1,1,2,2)),

Hi, I have been trying glht() from multcomp package and contrast() from contrast package to test a contrast that I am interested in. With the following simulated dataset (fixed effect "type" with 3 levels (b, m, t), and random effect "batch" of 4 levels, a randomized block design with interaction), sometimes both glht() and contrast() worked and gave nearly the same p values;

Dear R members, I tried to fit an lme model and to use the glht function of multcomp. However, the glht function gives me some errors when using weights=varPower(). The glht error makes sense as glht needs factor levels and the model works fine without weights=. Does anyone know a solution so I do not have to change the lme model? Thanks Sibylle --> works fine

Hi R users, I sent a message yesterday about NA in model estimates ( http://r.789695.n4.nabble.com/How-set-lm-to-don-t-return-NA-in-summary-td3722587.html). If I use aov() instead of lm() I get no NA in model estimates and I use gmodels::estimable() without problems. Ok! Now I'm performing a lot of contrasts and I need correcting for multiplicity. So, I can use multcomp::glht() for this.

Quick question about the usage of glht. I'm working with a data set from an experiment where the response is bounded at 0 whose variance increases with the mean, and is continuous. A Gamma error distribution with a log link seemed like the logical choice, and so I've modeled it as such. However, when I use glht to look for differences between groups, I get significant

Hi, I,ve a run a model like this mcrm<-glm(catroj~month,binomial) being catroj a binary response variable with two levels (infected and non infected) > anova(mcrm3,test="Chisq") Df Deviance Resid. Df Resid. Dev P(>|Chi|) NULL 520 149.81 mes 3 16.86 517 132.94 0.0007551 *** When I?m trying to do a post

Hi I have some experimental data where I have counts of the number of insects collected to different trap types rotated through 5 different location (variable -location), 4 different chemical attractants [A, B, C, D] were applied to the traps (variable - semio) and all were trialled at two different CO2 release rates [1, 2] (variable CO2) I also have a selection of continuous variables