Hi, I'm trying to set up AR(1) as a correlation structure in modeling some
data (attached file data.txt in text format) with lme, but have trouble
getting it to work.
Incent, Correctness, and Oppor are 3 categorical variables, Beta is a
response variable, and Time is an equally-spaced variable wit
For some unknown reason I stopped receiving any messages from the R-
help mailing list. See if this test gets through.
Thanks,
Gang
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http
With the example you provided, it seems both glht() and contrast()
work fine.
Based on my limited experience with contrast(), if you encounter such
an error message you just mentioned, check
> dat.lme$apVar
You might see something like this
[1] "Non-positive definite approximate variance
I'm running a categorical data analysis with a two-way design of
nominal by ordinal structure like the Political Ideology Example
(Table 9.5) in Agresti's book Categorical Data Analysis. The nominal
variable is Method while the ordinal variable is Quality (Bad,
Moderate, Good, Excellent). I
ts
>> x.new <- x[, -1] # delete "A"
>> x.new$FreqD <- counts # add new column
>> # print out unique entries
>> unique(x.new)
>B C D FreqD
> 1 B1 C1 D1 3
> 2 B2 C1 D1 3
> 4 B2 C2 D2 2
> 7 B1 C2 D2 2
>>
>
>
> On
Sorry the new column in DF2 should be called FreqD instead of FreqA.
How can I get DF2 with aggregate?
Gang
On Jan 7, 2008, at 2:38 PM, Gang Chen wrote:
> Yes, I misstated it when I said that I would keep B and C. I want to
> collapse column A, but count the frequency of D as a new col
C1 D243
B2 C2 D123
B2 C2 D243
Thanks,
Gang
On Jan 7, 2008, at 2:06 PM, Duncan Murdoch wrote:
> On 1/7/2008 1:28 PM, Gang Chen wrote:
>> I have a dataframe DF with 4 columns (variables) A, B, C, and D,
>> and want to create a new dataframe DF2 by keeping B and C
Thanks a lot! This is exactly what I wanted.
Gang
On Jan 5, 2008, at 2:20 PM, David Winsemius wrote:
> Gang Chen <[EMAIL PROTECTED]> wrote in
> news:[EMAIL PROTECTED]:
>
>> Suppose I have a two-way table of nominal category (party
>> affiliation) X ordinal c
I have a dataframe DF with 4 columns (variables) A, B, C, and D, and
want to create a new dataframe DF2 by keeping B and C in DF but
counting the frequency of D while collapsing A. I tried
by(DF$D, list(DF$B, DF$C), FUN=summary)
but this is not exactly what I want. What is a good way to do it
Suppose I have a two-way table of nominal category (party
affiliation) X ordinal category (political ideology):
party affiliation X (3 levels) - democratic, independent, and republic
political ideology Y (3 levels) - liberal, moderate, and conservative
The dependent variable is the frequency (o
Thanks a lot for all who've provided suggestions!
Gang
On Nov 15, 2007, at 5:09 PM, Duncan Murdoch wrote:
> On 11/15/2007 4:54 PM, Gang Chen wrote:
>> I want to identify whether a variable is character(0), but get
>> lost. For example, if I have
>> > dd&l
I want to identify whether a variable is character(0), but get lost.
For example, if I have
> dd<-character(0)
the following doesn't seem to serve as a good identifier:
> dd==character(0)
logical(0)
So how to detect character(0)?
Thanks,
Gang
__
Sorry to hijack this thread. I have a similar but slightly different
situation. Using the original poster's example, how to elegantly get
the mean of column V2 when column V1 is either A or C and F1 is 0?
Thanks,
Gang
On Nov 13, 2007, at 5:30 AM, Petr PIKAL wrote:
> Hi
>
> [EMAIL PROTECTED]
cc[n, ii, jj]
> }
> names(sublist) <- names(MyModel)
> result[[n]][[ii]] <- sublist
> }
> }
> str(result) # see what it looks like
>
>
>
> On Nov 8, 2007 6:31 PM, Gang Chen <[EMAIL PROTECTED]> wrote:
>> Thanks again for th
eplacement, so to see what the alternatives are, can you give an
> explicit example of what you would like as an outcome and then how you
> intend to use it.
>
> On Nov 8, 2007 5:19 PM, Gang Chen <[EMAIL PROTECTED]> wrote:
>> Thanks for the response!
>>
>> I want to create
gt;
> On Nov 8, 2007 4:51 PM, Gang Chen <[EMAIL PROTECTED]> wrote:
>> I have trouble creating an array of lists? For example, I want to do
>> something like this
>>
>> clist <- array(data=NA, dim=c(7, 2, 3));
>> for (n in 1:7) {
>>for (ii in 1:2) {
>
I have trouble creating an array of lists? For example, I want to do
something like this
clist <- array(data=NA, dim=c(7, 2, 3));
for (n in 1:7) {
for (ii in 1:2) {
for (jj in 1:3) {
if (cc[n, ii, jj] == "0") { clist[n, ii, ][[jj]] <- list(levels
(MyModel[,colnames(M
I want to create a list based on the information from a data.frame,
Model. So I tried the following:
MyList <- list(colnames(Model)[2] = levels(Model$(colnames(Model)[2])))
but it failed with an error:
Error: unexpected '=' in "list(colnames(Model)[2] ="
I have the following problems with th
actly a partial F test. So does
it mean that this only tests the average effect of those 3 terms? If
so, that would be slightly different from the partial F test I was
looking for, no?
Gang
> -G
>
>
>
> On Oct 30, 2007, at 5:26PM , Gang Chen wrote:
>
>> Dieter,
>>
>
0', 'IQ:I
(age^3) = 0'))
Error in FUN(newX[, i], ...) :
`param' has no names and does not match number of coefficients of
model. Unable to construct coefficient vector
Thanks,
Gang
On Oct 30, 2007, at 9:08 AM, Dieter Menne wrote:
> Gang Chen mail.nih.gov> write
>
> Of course this is yet another case where Lumley's principle (at
> least I
> think it's his) holds: if you have to use eval(parse(...)) rethink --
> there's a better way.
>
>
> Bert Gunter
> Genentech Nonclinical Statistics
>
>
> -Original
into MyFunc to make it
executable?
Gang
On Oct 29, 2007, at 5:42 PM, jim holtman wrote:
> Can you provide an example of your input and what you expect the
> output to be. You can always use 'as.numeric'.
>
> On 10/29/07, Gang Chen <[EMAIL PROTECTED]> wrote:
>> This
This must be very simple, but I'm stuck. I have a command line in R
defined as a variable of a string of characters. How can I convert
the variable so that I can execute it in R?
Really appreciate any help,
Gang
__
R-help@r-project.org mailing list
Suppose I have a mixed-effects model where yij is the jth sample for
the ith subject:
yij= beta0 + beta1(age) + beta2(age^2) + beta3(age^3) + beta4(IQ) +
beta5(IQ^2) + beta6(age*IQ) + beta7(age^2*IQ) + beta8(age^3 *IQ)
+random intercepti + eij
In R how can I get an F test against the nu
This is the command I want to execute with function contrast in
contrast package:
> contrast(MyObject, list(Trust="T"), list(Trust="U"));
As the two arguments with 'list' are defined as a string of
characters, contr:
> str(contr)
chr "list(Trust=\"T\"),list(Trust=\"U\")"
I would like to
Hi Liviu,
Thank you very much for the response! I knew that would work within
R, but was just wondering why
> R CMD INSTALL nlme
does not work on the shell terminal. Any clue?
Thanks,
Gang
On Oct 17, 2007, at 6:27 PM, Liviu Andronic wrote:
> On 10/17/07, Gang Chen <[EMAIL
fine:
> R CMD INSTALL lme4
Gang
On Oct 17, 2007, at 1:29 PM, Liviu Andronic wrote:
> On 10/17/07, Gang Chen <[EMAIL PROTECTED]> wrote:
>> Why does R CMD INSTALL work for some packages (e.g., lme4) but not
>> others (e.g., nlme)?
>
> If you don't provide any co
Why does R CMD INSTALL work for some packages (e.g., lme4) but not
others (e.g., nlme)?
Thanks,
Gang
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-gu
Yes, as.formula is the magic tool! Thanks a lot...
Gang
On Sep 19, 2007, at 2:00 PM, Vladimir Eremeev wrote:
>
> lme(as.formula(paste("y~",ww)),random=~1|subj,model)
>
>
> Gang Chen-3 wrote:
>>
>> I want to pass a predefined string ww ("fa*fb+fc")
I want to pass a predefined string ww ("fa*fb+fc") to function lme so
that I can run
> lme(y ~ fa*fb+fc, random = ~1|subj, model)
I've tried something like
> lme(y ~ paste(ww), random = ~1|subj, model)
and
> lme(y ~ sprintf(ww), random = ~1|subj, model)
but both give me the following
101 - 130 of 130 matches
Mail list logo