I'm running a mixed-effects model with lme in a big for loop of
20,000 iterations. It would take from 20 to 100 hours to finish a
typical analysis depending on the complexity of the model. However in
the model
Y = X*beta + Z*b + e
X and Z are always the same for all the iterations, and the
I want to check whether all the components of a vector (or an array)
are 0, and if they are I will skip the later computations. Of course
I can create a loop to go through all the components. However is
there an R function for this purpose more efficient than looping?
Thanks a lot,
Gang
Hi Lorenz,
I really appreciate your comments.
If I understand correctly, you want to include the interactions
between the random and fixed terms?
Yes that is exactly I wanted to model.
This is done like:
model.lme - lme(Beta ~ Trust*Sex*Freq,
random =
I'm trying to run a 3-way within-subject anova in lme with 3 fixed
factors (Trust, Sex, and Freq), but get stuck with handling the
random effects. As I want to include all the possible random effects
in the model, it would be something more or less equivalent to using aov
fit.aov -
I got a long list of error message repeating with the following 3
lines when running the loop at the end of this mail:
R(580,0xa000ed88) malloc: *** vm_allocate(size=327680) failed (error
code=3)
R(580,0xa000ed88) malloc: *** error: can't allocate region
R(580,0xa000ed88) malloc: *** set a
the memory usage keep accumulating? Does each error
message keep stored accumulatively in the buffer or something else?
Or something is wrong with the way I'm using 'try'?
Thanks,
Gang
On Aug 9, 2007, at 10:36 AM, Gang Chen wrote:
I got a long list of error message repeating with the following
I need some help on contrast testing with lme. It seems fit.contrast
in 'gmodels' package works fine for simple contrasting among levels
of a factor such as
fit.contrast(fit.lme, Trust, c(1,-1))
Estimate Std. Error t-value Pr(|t|)
Trust c=( 1 -1 ) -0.001442638
-
operator:
tag - 0
tryCatch(fit.lme - lme(Beta ~ Trust*Sex*Freq, random = ~1|Subj,
Model), error=function(err) tag - 1)
On Aug 6, 2007, at 6:55 PM, Seth Falcon wrote:
Gang Chen [EMAIL PROTECTED] writes:
I wanted something like this:
tag - 0;
tryCatch(fit.lme - lme(Beta ~ Trust*Sex
I'm trying to run a contrast with lme, but could not get it work:
anova(lme(Beta ~ Trust*Sex*Freq, random = ~1|Subj, Model), L=c
(TrustT:Sex:Freq=1, TrustU:Sex:Freq=-1))
Error in anova.lme(lme(Beta ~ Trust * Sex * Freq, random = ~1 | Subj, :
Effects TrustT:Sex:Freq, TrustU:Sex:Freq
I run a linear mixed-effects model in a loop
for (i in 1:N) {
fit.lme - lme(Y ~ X1*X2, random = ~1|subj, Model[i]);
}
As the data in some iterations are all (or most) 0's, leading to the
following error message from lme:
Error in chol((value + t(value))/2) : the leading minor of order 1
I run a linear mixed-effects model in a loop
for (i in 1:N) {
fit.lme - lme(Y ~ X1*X2, random = ~1|subj, Model[i]);
}
As the data in some iterations are all (or most) 0's, leading to the
following error message from lme:
Error in chol((value + t(value))/2) : the leading minor of order 1
with 'err' or just return 0 as above)
--- Gang Chen [EMAIL PROTECTED] wrote:
I run a linear mixed-effects model in a loop
for (i in 1:N) {
fit.lme - lme(Y ~ X1*X2, random = ~1|subj, Model[i]);
}
As the data in some iterations are all (or most) 0's, leading to the
following error message
, Stephen Tucker wrote:
?try
or
?tryCatch
http://www.maths.lth.se/help/R/ExceptionHandlingInR/
for example...
tryCatch(lme(Y ~ X1*X2, random = ~1|subj, Model[i]),
error=function(err) return(0))
(you can do something with 'err' or just return 0 as above)
--- Gang Chen [EMAIL
I have a mixed balanced ANOVA design with a between-subject factor
(Grp) and a within-subject factor (Rsp). When I tried the following
two commands which I thought are equivalent,
fit.lme - lme(Beta ~ Grp*Rsp, random = ~1|Subj, Model);
fit.aov - aov(Beta ~ Rsp*Grp+Error(Subj/Rsp)+Grp,
?
Thanks,
Gang
On Aug 3, 2007, at 3:52 PM, Peter Dalgaard wrote:
Gang Chen wrote:
I have a mixed balanced ANOVA design with a between-subject
factor (Grp) and a within-subject factor (Rsp). When I tried the
following two commands which I thought are equivalent,
fit.lme - lme(Beta ~ Grp
corrected)
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Gang Chen
Sent: Friday, August 03, 2007 4:01 PM
To: Peter Dalgaard
Cc: r-help@stat.math.ethz.ch
Subject: Re: [R] lme and aov
Thanks for the response!
It is indeed a balanced design. The results
Subj as factor.
This is exactly the problem I had: Model$Subj was not a factor! Now
they converge. A lesson well learned.
Thanks a lot for the help,
Gang
On Aug 3, 2007, at 4:53 PM, Peter Dalgaard wrote:
Gang Chen wrote:
Thanks a lot for clarification! I just started to learn
Thanks a lot for the response and explanation, Lorenz! It seems that
I can run the analysis with 'lm' without treating subject as a random
factor
lm(Resp ~ Group*Cov, TestData)
Thanks,
Gang
On Aug 2, 2007, at 5:09 AM, [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
I do not think anyone has
Sorry about this basic question. After reading a table,
Model=read.table(ModelMat.txt, header=T)
I want to get access to each entry in the table Model. However, if I do
Model[1,1]
I get the following,
[1] A
Levels: A B C
My question is, how can I just get the entry A without the 2nd line
I would like to run a regression analysis without a constant
(intercept) or a special one-way within-subject (repeated-measures)
ANOVA. I'm not sure if the following command lines are correct or not:
m1 - lme(Resp ~ Cond - 1, random = ~ Cond - 1 | Subj, TestData)
or,
m2 - lmer(Resp ~ Cond
I'm trying to run a simple one-way ANCOVA with the lmer function in R
package lme4, but have encountered some conceptual problem. The data
file MyData.txt is like this:
Group Subj Cov Resp
A 1 3.90 4.05
A2 4.05 4.25
A3 4.25 3.60
A4 3.60
Based on the examples I've seen in using statistical analysis
packages such as lmer, it seems that people usually tabulate all the
input data into one file with the first line indicating the variable
names (or labels), and then read the file inside R. However, in my
case I can't do that
it
in a loop?
Thanks,
Gang
On Tue, 24 Jul 2007, Gang Chen wrote:
Based on the examples I've seen in using statistical analysis
packages such as lmer, it seems that people usually tabulate all the
input data into one file with the first line indicating the variable
names (or labels
Hi all,
I am a newbie of R, and have a very simple question. When I run the
following t test
result = t.test(weight ~ group)
If I want to get the t value, I can do
tvalue = result$statistic
with tvalue being the following:
t
1.191260
However, if I just want the number 1.191260, not the
Thank you very much!
Gang
On Jul 17, 2007, at 3:57 PM, Duncan Murdoch wrote:
On 7/17/2007 3:36 PM, Gang Chen wrote:
Hi all,
I am a newbie of R, and have a very simple question. When I run
the following t test
result = t.test(weight ~ group)
If I want to get the t value, I can do
Hi all,
How can I invoke an operating system command in R? I mean something
like exclamation mark (!) inside Matlab.
Thanks,
Gang
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
Hi,
I am trying to run some path analysis with Dr. Fox's sem package. The
number N in sem(ram, S, N) is supposed to be the total number of
observations, right? However, in my situation the effective number of
degrees of freedom for each observed variable is estimated by some
in library(sem) : 'sem' is not a valid package -- installed
2.0.0?
Why is this?
Thank you,
Gang Chen
[[alternative HTML version deleted]]
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read
28 matches
Mail list logo
