On Mon, 23 Feb 2004, Jason Turner wrote:
has anyone got some advise about loading MiceR,
...
I note from the web page that MiceR is described as a package for Unix
and I am running R on Windows XP. Does it make a difference to the
package?
Rather a lot. You can
1) compile it
Hi,
Martin Maechler wrote:
Back from my vacation, I haven't seen an R-help answer on this
(Christian, where have you been ? ;-)
(Uh, I missed this one. Too much spam?)
I would add information based criteria (AIC, BIC and so on)
together with a normal mixture model (implemented in
Dear All,
I would like to ask how to customize the graph corresponding to a
procrustes analysis.
I have to distance matrices, that I transform to two set of coordinates
by means of muti dimensional scaling:
library(mva)
c1-cmdscale(mat.dist1)
c2-cmdscale(mat.dist2)
I vant to rotate c2 on c1,
Dear Lisa,
as far as I know, there is no implementation of dendrogram ultrametrics,
but it could be easily computed from the output of hclust (package mva)
or agnes (package cluster) in the following manner:
1) Generate the desired dendrogram by hclust or agnes.
2) Generate partitions for all
Hello all!
When you are working with a little more complicated models in
SAS PROC MIXED, you often use the /ddfm=satterth call to make sure
the df decomposition is done the best way possible.
Running the same models in lme, without any special calls, results
in warning messages about the df
Federico Calboli wrote:
Dear All,
I would like to ask how to customize the graph corresponding to a
procrustes analysis.
SNIP
The object returned by procrustes() contains Yrot and X, the rotated
matrix and the target matrix respectively. You could plot both of these
using standard plotting
The `plot.procrustes' function really should be more user-friendly and
flexible. You should contact its author and ask for amendments.
I honestly though my problem was too trival to bother the author in
person, and I thought that getting an answer would leave it in a public
database, as future
Dear all,
When a conventional kriging and then a contour plot is limited with a polygon (as
possible with krige.conv and contour.kriging), the
polygon border is displayed in black by default.
kc-krige.conv(CZdata,loc=pred.grid,borders=czpoly,krige=krige.control(obj.m=ls))
You are not allowed to post to this mailing list, and your message has
been automatically rejected. If you think that your messages are
being rejected in error, contact the mailing list owner at
[EMAIL PROTECTED]---BeginMessage---
that is bad
---End Message---
Qin -
If you do str(pred), I think you will find that pred is
now a factor with 276 levels. to convert it back into an
ordinary numeric vector, do
values - as.numeric(as.character(pred))
- tom blackwell - u michigan medical school - ann arbor -
On Mon, 23 Feb 2004, Qin Liu wrote:
Sounds like you are trying to do k-NN regression (i.e., predict numeric
variable given others) with 12 predictors, some of which are categorical.
If so, you're quite possibly out of luck. k-NN methods depend on distances,
and you need to have a distance that make sense for your problem and works
Greetings.
One for the developers I guess... I am having problems in using a
(non-recommended) Fortran compiler (Salford ftn95 Windoze), and
the crashes do seem to be associated with the ftn95-dervived DLL
changing the Floating Point Control Word. The compiler people are
suggesting (and I
I hope that the mailing list is the correct forum for the question below. I
have trouble calling functions nlsList and nlme from
another function. Any help would be greatly appreciated.
Jens Praestgaard
Human Genome Sciences
Rockville MD.
I have a data set v with two components, v$mixeddat and
For what it's worth my forthcoming book, An Introduction to Modern Bayesian
Econometrics, Blackwells, May 2004,
makes extensive use of R.
Tony Lancaster
[[alternative HTML version deleted]]
__
[EMAIL PROTECTED] mailing list
On Mon, 23 Feb 2004 13:37:21 +, Steve Roberts
[EMAIL PROTECTED] wrote :
Greetings.
One for the developers I guess... I am having problems in using a
(non-recommended) Fortran compiler (Salford ftn95 Windoze), and
the crashes do seem to be associated with the ftn95-dervived DLL
changing
Dear Patrick
border=NULL should work but there is a bug.
I'm fixing this right now and a new version will be available later today
P.J.
On Mon, 23 Feb 2004, Patrick Giraudoux wrote:
Dear all,
When a conventional kriging and then a contour plot is limited with a polygon (as
possible with
Thanks a lot for the answer!
Now, I only have the last one left - How do I get round it?
I knew about the missing cells in the design, but didn´t know how lme
would react on them.
In this case, I can remove the water:temp term, but how can I be sure
that this is the right thing to do?
Is the lm
If you want to try to get the same answers as PROC MIXED, I
suggest you try to figure out how SAS codes interactions and which ones
it retains. Then you can try code those manually and include them as
separate explanatory variables, e.g., I((water==2)(temp==110)). You
could work this
Hello, I try to compute means,sd,n from a variable y according to some
categoricals g1,g2... I wrote the following function:
msd-function(y,...) {
print(match.call())
funy-function(x) list(mean(x,na.rm=T),sd(x,na.rm=T),table(is.na(x))[1])
gg - list(...)
aa-by(y,gg,funy)
Dear R-listers,
I got an error when I try to plot two grouped data into a single
win.metafile device:
library(lattice)
trellis.device(device=win.metafile,color=F,filename=Profiles-Var1.wmf)
par(mfrow=c(1,2))
# First plot
xyplot(log(v)~t|id,data=Con.20,
main=Group A: Control, Var-1,
Why not just do:
tapply(y, list(g1, g2, ...), function(x) c(mean=mean(x, na.rm=TRUE),
sd=sd(x, na.rm=TRUE), n=sum(!is.na(x
??
Andy
From: r.ghezzo
Hello, I try to compute means,sd,n from a variable y
according to some
categoricals g1,g2... I wrote the following function:
On Monday 23 February 2004 09:57, Rodrigo Abt wrote:
Dear R-listers,
I got an error when I try to plot two grouped data into a single
win.metafile device:
library(lattice)
trellis.device(device=win.metafile,color=F,filename=Profiles-Var1.wmf
) par(mfrow=c(1,2))
par() settings have
You cannot use par(), which is part of the base graphics system, with
xyplot(), which is part of the lattice/grid system. If you want to
combine base-graphics with lattice/grid you need to use the `gridBase'
package on CRAN.
-roger
Rodrigo Abt wrote:
Dear R-listers,
I got an error when I
In a reply to my inquiry, a respondent offered the following simulation code:
lens - rpois(10, 3)
V - numeric(10)
for(i in 1:length(lens)) V[i] - sum(rnorm(lens[i]))
quantile(V, c(.95, .99))
The code worked exactly the way I wanted. I had an Excel model that does the same
thing.
CG Pettersson [EMAIL PROTECTED] writes:
Thanks a lot for the answer!
Now, I only have the last one left - How do I get round it?
I knew about the missing cells in the design, but didn´t know how lme
would react on them.
In this case, I can remove the water:temp term, but how can I be sure
I apologize in advance for posting a question not related to R.
I need references to papers that use multiple linear regression in an industrial
application and also in a biological experiment.
This is aimed to biologists/chemists (non-statisticans) so if anyone has written a
brilliant paper
Hello,
I created a simple histogram with:
myHist-hist(myData)
the object myHist now has two arrays (among the others):
myHist$mids
myHist$counts
Since myHist$counts contains some 0, and I want to calculate the linear fit among
myHist$mids and log(myHist$counts), I want remove the elements of
Repeatedly people have asked how to pass arguments by reference in R.
Now package ref is on CRAN, which provides two referencing methods:
Functions ref(), deref() and friends conveniently allow to pass parameters
by reference instead of by value. This can be useful in memory critical
Doug's xtabs suggests to me that the following might be
estimable, with data.$Temp - as.numeric(as.character(data.$temp))
water*(Temp+I(Temp^2))
It looks like it should be estimable in lm, and depending on the
noise model, it should also be estimable in lme. ???
hope this
Chemists refer to regression as QSAR so googling for that
should find you lots of examples.
Date: Mon, 23 Feb 2004 18:36:47 +0100
From: Andersson, Henrik [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Subject: [R] Reference to use of MLR in industry and biology
I apologize in advance
You can create a new data frame and subset it:
myHist.df - data.frame(counts = myHist$counts, mids = myHist$mids )
myHist.df - subset( myHist, counts 0 )
lm( mids ~ log(counts), data = myHist.df )
---
Date: Mon, 23 Feb 2004 18:40:22 +0100 (CET)
From: =?iso-8859-1?q?Fulvio=20Copex?=
Hi, I am a new user of R. I am trying to use R to do some nonlinear modeling. However,
I can not find how to get the parameters's value back.
Here is the code in S-plus:
ycos1.nl - nls(y ~ m +ac*cos(2*pi*f*t) +as*sin(2*pi*f*t), start=nl.st, trace=T)
## store model info after satisfactory
I know this:
library(date)
x=1979-04-04
try=as.date(x, ymd)
print(try)
[1] 4Apr79
and that `x' here has to be a string, e.g.:
x=1979-04-04
print(x)
[1] 1971
I'm stuck in reading from a file. I say:
A - read.table(file=try)
print(A)
V1 V2
1
Hi,
Is there any existing functions to open an HTTP
connection and HTTP POST some R data?
like
form method=post enctype=multipart/form-data
etc.
/form
but within R.
Thanks,
=
Nathan Whitehouse
Statistics/Programming
Baylor College of Medicine
Houston, TX, USA
[EMAIL
On Tue, 24 Feb 2004, Ajay Shah wrote:
I know this:
library(date)
x=1979-04-04
try=as.date(x, ymd)
print(try)
[1] 4Apr79
and that `x' here has to be a string, e.g.:
x=1979-04-04
print(x)
[1] 1971
I'm stuck in reading from a file. I say:
A -
Ajay Shah wrote:
I know this:
library(date)
x=1979-04-04
try=as.date(x, ymd)
print(try)
[1] 4Apr79
and that `x' here has to be a string, e.g.:
x=1979-04-04
print(x)
[1] 1971
I'm stuck in reading from a file. I say:
A - read.table(file=try)
print(A)
At 1:37 AM +0530 2/24/04, Ajay Shah wrote:
I know this:
library(date)
x=1979-04-04
try=as.date(x, ymd)
print(try)
[1] 4Apr79
and that `x' here has to be a string, e.g.:
x=1979-04-04
print(x)
[1] 1971
I'm stuck in reading from a file. I say:
A - read.table(file=try)
Duan Zhigang [EMAIL PROTECTED] writes:
Hi, I am a new user of R. I am trying to use R to do some nonlinear
modeling. However, I can not find how to get the parameters's value
back.
Here is the code in S-plus:
ycos1.nl - nls(y ~ m +ac*cos(2*pi*f*t) +as*sin(2*pi*f*t), start=nl.st, trace=T)
Are you sure? I just checked on CRAN (US mirror) and it says:
VR:
Functions and datasets to support Venables and Ripley, 'Modern Applied
Statistics with S' (4th edition). Bundle of: MASS class nnet spatial
Version: 7.1-14
Priority: recommended
Depends: R (= 1.8.0), lattice, nlme (= 3.1-40),
Perez Martin, Agustin wrote:
DeaR useRs:
I am looking for a function which fits a multinomial model and in Baron´s
page I find the function multinom in package nnet but this package is
deprecated.
Really? What do you mean with deprecated?
I suppose that this function is now in other package but
`Recommended' not `deprecated'.
nnet is part of the VR bundle that should be in all complete R
installations. Did you actually look in yours?
On Mon, 23 Feb 2004, Perez Martin, Agustin wrote:
DeaR useRs:
I am looking for a function which fits a multinomial model and in Baron´s
page I
Frank == Frank E Harrell [EMAIL PROTECTED] writes:
Also, has anyone tried creating a Word document using
OpenOffice with figures imported from R?
I've tried that. It works fine with openoffice, but not with
word. For example, create a postscript file with R like this:
If both functions are univariate, you can define the difference between the
two as the objective function and use uniroot() to find the intersection
(where the difference is 0). If the functions are multivariate, you can use
optim() to minimize the squared or absolute difference between the two
Hi Fellows from R-Help List!
My questions are basic since i an new with R. I am very acquainted with Matlab
Gauss (the compentence, I guess). Anyhow,
(1) I am trying to get R execute comands made or built as text, so that one can
feed a particular option with many variations coming from a text
Hi,
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
[EMAIL PROTECTED]
Sent: Tuesday, February 24, 2004 12:34 PM
To: [EMAIL PROTECTED]
Subject: [R] (2) Questions
Hi Fellows from R-Help List!
My questions are basic since i an new with R. I
From: [EMAIL PROTECTED]
Hi Fellows from R-Help List!
My questions are basic since i an new with R. I am very
acquainted with Matlab
Gauss (the compentence, I guess). Anyhow,
(1) I am trying to get R execute comands made or built as
text, so that one can
feed a particular option
On Mon, 2004-02-23 at 18:53, li xian wrote:
How to get the sum of the diag of matrix?
Thanks!
If 'm' is your matrix:
sum(diag(m))
See ?sum and ?diag, the latter of which will extract the diagonal of the
matrix.
Be sure to read the help for diag() fully for some of the usage caveats.
HTH,
A - array(1:4, dim=c(2,2))
sum(diag(A))
[1] 5
Is that what you want?
spencer graves
li xian wrote:
How to get the sum of the diag of matrix?
Thanks!
-
[[alternative HTML version deleted]]
__
[EMAIL
sum(diag(X))
On Mon, 23 Feb 2004, li xian wrote:
How to get the sum of the diag of matrix?
Thanks!
-
[[alternative HTML version deleted]]
__
[EMAIL PROTECTED] mailing list
Let QR be such that sqrt(D)X = QR. Then letting
solve(...) denote the inverse of ... we have
X = solve(sqrt(D))QR which is of the form ZR
and Z has the desired weighted orthoginality
property.
Since D is diagonal, solve(sqrt(D)) equals
diag(1/sqrt(diag(D))) so we get this for Z:
In my .Snw file:
=
fyle - choose.files()
fyle
f - count.fields(fyle, sep = \t)
f
@
and in the .tex file:
\begin{Schunk}
\begin{Sinput}
fyle - choose.files()
fyle
\end{Sinput}
\begin{Soutput}
[1] C:\\Files\\Data\\Cars03\\Rover.txt
\end{Soutput}
\begin{Sinput}
f - count.fields(fyle, sep =)
G'day,
I'm fitting a simple one-way nested anova and would like to obtain standard
errors or confidence limits. I'm using se.contrast to try to get se for
the contrast between the two levels of the main effect, but I get an error
message (see below). What is going on?
Don
G'day,
I'm fitting a simple one-way nested anova and would like to obtain standard
errors or confidence limits. I'm using se.contrast to try to get se for
the contrast between the two levels of the main effect, but I get an error
message (see below). What is going on?
Don
53 matches
Mail list logo