### [R] Deviance function in regression trees

```Hello all. I have heard over and over that CART and its various tree-like
brethren are non-parametric techniques. When I read the chapter in
Chambers and Hastie on tree-based models it states that tree-based models
can be generalized (GTMs) in a manner similar to GLMs by specifying a
different deviance function to distributions other than the gaussian error
distribution ( section 9.4.3).  I have an application in which the response
variable is a continuous variable representing tree counts within a unit
area and thus would be best described by a poisson distribution. The error
distribution for this data is not gaussian. If this is the case, will the
gaussian error distribution used in most regression tree packages, be
appropriate? Are there ways to specify the error distribution in R or Should
I log transform the response variable?  If the specification of error
distribution in regression trees is important, than are these techniques
truly  non-parametric. Thanks for your inputs.

Solomon Dobrowski
Tahoe Environmental Research Center (TERC)
John Muir Institute of the Environment
University of California, Davis
530 754 9354

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Legend box line thickness

```I am making a plot and am merely trying to increase the line thickness, or
width, of the box drawn around the legend.  The help page on 'legend' was of
no use.  Does anyone have any ideas?

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Type II and III sum of square in Anova (R, car package)

```I cannot resist a very brief entry into this old and seemingly
immortal issue, but I will be very brief, I promise!

Amasco Miralisus suggests:

As I understood form R FAQ, there is disagreement among Statisticians
which SS to use

(http://cran.r-project.org/doc/FAQ/R-FAQ.html#Why-does-the-output-from-a
nova_0028_0029-depend-on-the-order-of-factors-in-the-model_003f).

To let this go is to concede way too much.  The 'disagreement' is
really over whether this is a sensible question to ask in the first
place.  One side of the debate suggests that the real question is what
hypotheses does it make sense to test and within what outer
hypotheses.  Settle that question and no issue on types of sums of
squares arises.

This is often a hard question to get your head around, and the
attraction of offering a variety of 'types of sums of squares' holds
out the false hope that perhaps you don't need to do so.  The bad
news is that for good science and good decision making, you do.

Bill Venables.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Help with Functions

```Hi

check out R-and-octave.txt on the contributed docs section of CRAN.

HTH

Robin

On 28 Aug 2006, at 17:50, Lord Tyranus wrote:

Hello wizards, I need to convert the following functions (prestd,
poststd, prepca)   of matlab to R. Does Somebody knows how to do it. A
description of the functions is:

prestd preprocesses the network training set by normalizing the inputs
and targets so that they have means of zero and standard deviations of
1.

poststd postprocesses the network training set which was preprocessed
by prestd. It converts the data back into unnormalized units.

prepca preprocesses the network input training set by applying a
principal component analysis. This analysis transforms the input data
so that the elements of the input vector set will be uncorrelated. In
addition, the size of the input vectors may be reduced by retaining
only those components which contribute more than a specified fraction
(min_frac) of the total variation in the data set.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
guide.html
and provide commented, minimal, self-contained, reproducible code.

--
Robin Hankin
Uncertainty Analyst
National Oceanography Centre, Southampton
European Way, Southampton SO14 3ZH, UK
tel  023-8059-7743

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Deviance function in regression trees

```The short answer is that a Poisson distribution is a discrete
distribution: if that is appropriate to your data the rpart function (in
the package of that name) has a suitable option.

On Mon, 28 Aug 2006, Solomon Dobrowski wrote:

Hello all. I have heard over and over that CART and its various tree-like
brethren are non-parametric techniques.  When I read the chapter in
Chambers and Hastie on tree-based models it states that tree-based models
can be generalized (GTMs) in a manner similar to GLMs by specifying a
different deviance function to distributions other than the gaussian error
distribution ( section 9.4.3).  I have an application in which the response
variable is a continuous variable representing tree counts within a unit
area and thus would be best described by a poisson distribution. The error
distribution for this data is not gaussian. If this is the case, will the
gaussian error distribution used in most regression tree packages, be
appropriate? Are there ways to specify the error distribution in R or Should
I log transform the response variable?  If the specification of error
distribution in regression trees is important, than are these techniques
truly  non-parametric. Thanks for your inputs.

Solomon Dobrowski
Tahoe Environmental Research Center (TERC)
John Muir Institute of the Environment
University of California, Davis
530 754 9354

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

--
Brian D. Ripley,  [EMAIL PROTECTED]
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel:  +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax:  +44 1865 272595

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Firefox extension fo R Site Search

```ecatchpole a écrit :
Romain Francois wrote on 08/28/2006 04:23 PM:

=
Search for your R routines directly from Firefox!
=

When I try to search I get the Firefox error

TypeError: document.getElementById(currentPackage + : +
currentHelpPage) has no properties

Help!  (Screen shot emailed separately to Romain.)

Ted
Hi Ted,

Good catch. I'll fix it today.
Did you try to search for something ?

BTW, I'll set up a page in the developers section of the wiki to discuss
(problems with) this project. To appear soon.

Cheers,

Romain

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] standardized partial regression coefficients

```Yuval Sapir wrote:
Hi all,
I am relatively new to R, so let me know if I missed something trivial.
I want to perform path-analysis with at least three levels of explaining
variables (i.e., A affects B that affects C). I perform multiple
regression for each set of variables, and get estimates for the partial
coefficients and their SE. How can I translate it to standardized
partial regression coefficients (beta in SPSS)?
Thanks
Yuval

RSiteSearch(standardized coefficients) is helpful.  For example:

http://finzi.psych.upenn.edu/R/Rhelp02a/archive/68367.html
http://finzi.psych.upenn.edu/R/Rhelp02a/archive/6038.html

You also might consider the sem package by John Fox with its
standardized.coefficients function.

http://cran.r-project.org/doc/packages/sem.pdf

--
Chuck Cleland, Ph.D.
NDRI, Inc.
71 West 23rd Street, 8th floor
New York, NY 10010
tel: (212) 845-4495 (Tu, Th)
tel: (732) 512-0171 (M, W, F)
fax: (917) 438-0894

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] EOF and CCA analysis

```Hello!
I would like to do Empirical Orthogonal functions and
Canonical correlation analysis on satellite data. My
matrices are going to be very big (more than 10,000
locations). Are there limitations in R regarding the
size of the matrix?
thanks a lot
regards
Isidora

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] AffyChip Background Analysis

```Dear list,

I want to analyse some HG133Plus2.0 Affymetrix chips.
The first thing I intent to do, is just to perform a background correction
(mas, rma, gcrma,...) with the cel files.
Then I want to take a look at the files.
How can I do this ?

Gunther

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Remove empty list from list - remove only one row and make as matrix

```Dear all,
After I work around, I found in my list of data with only one row which
need to remove or make it as matrix.
Here I write again my other toy example:
x - list(matrix(1:20, 5, 4),matrix(1:20, 5, 4),matrix(1:20, 5,
4),matrix(1:20, 5, 4),matrix(1:20, 5, 4))
y - list(c(1, -1, -1, 1, 1),c(1, -1, -1, -1, -1),c(1, 1, 1, 1, 1),c(1,
1, -1, 1, -1),c(-1, -1, -1, -1, -1))
## Thanks to Gabor Grothendieck for this trick.
## SIMPLIFY? SIMPLIFY  simplify
xy.list - mapply(cbind, x, y, SIMPLIFY=FALSE)

point.class -
t(cbind(c(10,20,15,4,-1),c(21,10,15,34,-1),c(11,13,6,3,1),c(7,5,5,2,1),c(8,9,5,12,-1)))
class.diffsame - points.neighb(as.matrix(point.class), xy.list, 5)
pd.class - points.diff(class.diffsame,xy.list)

nc.test - vector(list,length(pd.class))
for (i in 1:length(pd.class)) {
nc.test[[i]] - pd.class[[i]]\$point.diff
}
nc.test

# delete null/empty entries in a list
dff - delete.NULLs(nc.test)
dff; str(dff)
dff
[[1]]
[,1] [,2] [,3] [,4] [,5]
[1,]16   11   161
[2,]49   14   191
[3,]5   10   15   201

[[2]]
[1]  1  6 11 16  1

[[3]]
[,1] [,2] [,3] [,4] [,5]
[1,]38   13   18   -1
[2,]5   10   15   20   -1

lapply(dff, nrow)
lapply(dff, nrow)
[[1]]
[1] 3

[[2]]
NULL

[[3]]
[1] 2

#I can use
#dff[unlist(lapply(dff, nrow) == 1)] #2,3, etc

I have two questions here:
a. I need to remove dff[[2]]
b. How to make it as matrix (in list). I mean the result something like

[[1]]
[,1] [,2] [,3] [,4] [,5]
[1,]16   11   161
[2,]49   14   191
[3,]5   10   15   201

[[2]]
[,1] [,2] [,3] [,4] [,5]
[1]1611   161

[[3]]
[,1] [,2] [,3] [,4] [,5]
[1,]38   13   18   -1
[2,]5   10   15   20   -1

points.neighb - function(p.class, list.nc, class.col) {
ntuples - nrow(p.class)
instvec - vector(list,length=ntuples)
for (i in 1:ntuples) {
# Thanks to Petr Pikal for this trick
instvec[[i]]\$class.diff - (p.class[i,class.col] -
list.nc[[i]][,class.col])!=0
instvec[[i]]\$class.same - (p.class[i,class.col] -
list.nc[[i]][,class.col])==0
}
instvec
}

points.diff - function(p.class, list.nc) {
ntuples - length(list.nc)
instvec - vector(list,ntuples)
for (i in 1:ntuples) {
instvec[[i]]\$point.diff - list.nc[[i]][p.class[[i]]\$class.diff,]
instvec[[i]]\$point.same - list.nc[[i]][p.class[[i]]\$class.same,]
}
instvec
}

# Thanks to Jim Holtman for this trick
delete.NULLs  -  function(x.list){
x.list[unlist(lapply(x.list, length) != 0)]
}

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Type II and III sum of square in Anova (R, car package)

```Amasco,
In general it is dangerous to attempt to interpret a main effect that
is included in an interaction, regardless of wether or not the
interaction is significant. If you want to make a valid inference about
a main effect it is safest to do so after dropping any interaction that
contains the main effect. Since you would not want to drop a significant
interaction, you should not try to interpret a main effect in the
presence of a significant interaction that contains the main effect. If
the interaction is not significant drop the interaction, re-run the
model and then look at the main effect.
John

John Sorkin M.D., Ph.D.
Chief, Biostatistics and Informatics
Baltimore VA Medical Center GRECC,
University of Maryland School of Medicine Claude D. Pepper OAIC,
University of Maryland Clinical Nutrition Research Unit, and
Baltimore VA Center Stroke of Excellence

University of Maryland School of Medicine
Division of Gerontology
Baltimore VA Medical Center
10 North Greene Street
GRECC (BT/18/GR)
Baltimore, MD 21201-1524

(Phone) 410-605-7119
(Fax) 410-605-7913 (Please call phone number above prior to faxing)
[EMAIL PROTECTED]

Amasco Miralisus [EMAIL PROTECTED] 8/28/2006 3:20 PM
Hello,

First of all, I would like to thank everybody who answered my
question. Every post has added something to my knowledge of the topic.
I now know why Type III SS are so questionable.

As I understood form R FAQ, there is disagreement among Statisticians
which SS to use
(http://cran.r-project.org/doc/FAQ/R-FAQ.html#Why-does-the-output-from-anova_0028_0029-depend-on-the-order-of-factors-in-the-model_003f).
However, most commercial statistical packages use Type III as the
default (with orthogonal contrasts), just as STATISTICA, from which I
am currently trying to migrate to R. This was probably was done for
the convenience of end-users who are not very experienced in
theoretical statistics.

I am aware that the same result could be produced using the standard
anova() function with Type I sequential SS, supplemented by drop1()
function, but this approach will look quite complicated for persons
without any substantial background in statistics, like no-math
students. I would prefer easier way, possibly more universal, though
also probably more for dummies :) If am not mistaken, car package by
John Fox with his nice Anova() function is the reasonable alternative
for any, who wish to simply perform quick statistical analysis,
without afraid to mess something with model fitting. Of course
orthogonal contrasts have to be specified (for example contr.sum) in
case of Type III SS.

Therefore, I would like to reformulate my questions, to make it easier

1. The first question related to answer by Professor Brian Ripley: Did
I understood correctly from the advised paper (Bill Venables'
'exegeses' paper) that there is not much sense to test main effects if
the interaction is significant?

2. If I understood the post by John Fox correctly, I could safely use
Anova(.,type=III) function from car for ANOVA analyses in R, both
for balanced and unbalanced designs? Of course providing the model was
fitted with orthogonal contrasts. Something like below:
mod - aov(response ~ factor1 * factor2, data=mydata,
contrasts=list(factor1=contr.sum,
factor2=contr.sum))
Anova(mod, type=III)

It was also said in most of your posts that the decision of which of
Type of SS to use has to be done on the basis of the hypothesis we
want to test. Therefore, let's assume that I would like to test the
significance of both factors, and if some of them significant, I plan
to use post-hoc tests to explore difference(s) between levels of this
significant factor(s).

On 8/27/06, John Fox [EMAIL PROTECTED] wrote:
Dear Amasco,

A complete explanation of the issues that you raise is awkward in an
email,
Applied
Regression Analysis, Linear Models, and Related Methods (Sage, 1997)
has a
detailed discussion.

(1) In balanced designs, so-called Type I, II, and III sums of
squares
are identical. If the STATA manual says that Type II tests are only
appropriate in balanced designs, then that doesn't make a whole lot
of sense
(unless one believes that Type-II tests are nonsense, which is not
the
case).

(2) One should concentrate not directly on different types of sums
of
squares, but on the hypotheses to be tested. Sums of squares and
F-tests
should follow from the hypotheses. Type-II and Type-III tests (if the
latter
are properly formulated) test hypotheses that are reasonably
construed as
tests of main effects and interactions in unbalanced designs. In
unbalanced
designs, Type-I sums of squares usually test hypotheses of interest
only by
accident.

(3) Type-II sums of squares are constructed obeying the principle of
marginality, so the kinds of contrasts employed to represent factors
are
irrelevant to the sums of squares produced. ```

### Re: [R] screen resolution effects on graphics

```In thinking about this some more Romain's idea of using
the browser to get this information can be done like this.
This solution only works with Internet Explorer:

library(RDCOMClient)
ie - COMCreate(InternetExplorer.Application)
width - ie[[document]][[ParentWindow]][[screen]][[width]]
height - ie[[document]][[ParentWindow]][[screen]][[height]]
ie\$Quit()

On 8/28/06, Gabor Grothendieck [EMAIL PROTECTED] wrote:
I think there are problems with browseURL on Windows XP with IE.
Put the file in screen.htm and then run:

shell.exec(screen.htm)

it will come up -- it will block it and you will have to click to
unblock it; however,
that will interfere with any automatic procedure.

On 8/28/06, Charles Annis, P.E.
[EMAIL PROTECTED] wrote:
Romain:

a - tempfile()
cat('htmlscript type=text/javascript document.write(screen.width) ;
/script/html', file=a)
browseURL(a)

The object a was created, but  no browser opened.

ls()
[1] a
a
[1] C:\\DOCUME~1\\CHARLE~1\\LOCALS~1\\Temp\\RtmpRgWrqb\\file678418be

Charles Annis, P.E.

[EMAIL PROTECTED]
phone: 561-352-9699
eFax:  614-455-3265
http://www.StatisticalEngineering.com

-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Romain Francois
Sent: Monday, August 28, 2006 2:30 PM
To: Prof Brian Ripley
Cc: Charles Annis, P.E.; r-help@stat.math.ethz.ch
Subject: Re: [R] screen resolution effects on graphics

Prof Brian Ripley a écrit :
On Mon, 28 Aug 2006, Charles Annis, P.E. wrote:

Greetings, R-Citizens:

I have the good fortune of working with a 19 1280 X 1024 pixel monitor.
My

(Similar to our student lab has used for many years.)

R-code produces nice-looking graphics on this machine but the same code
results in crowded plots on an older machine with 800 X 600 resolution.
In
hindsight this seems obvious, but I didn't anticipate it.

It is not obvious to me: I have never experienced it.  What OS and
graphics device is this?

Almost all of R's graphics is independent of the screen resolution (the
exception being the bitmapped devices such as jpeg), with things sized in
inches or points. My machines are 1600x1200 (apart from 1280x800 on my
laptop), so I meet a considerable reduction when using a computer
projector, and my plots do not look crowded.

However, one issue is when the OS has a seriously incorrect setting for
the screen resolution and so does not give the sizes asked for by R.  We
have seen that on both Linux and Windows, and the windows() device has
arguments to set the correct values.  (On X11 you should be able to set
this in Xconfig files.)

If this is Windows, check carefully the description of the initial screen
size in ?windows.  That can have unexpected effects on physically small
screens.

At one time the X11() device was set up to assume 75dpi unless the
reported resolution was 100+/-0.5dpi.  My then monitor reported 99.2 dpi
and so things came out at 3/4 of the intended size.  We fixed that quite a

while back.

My code will be used on machines with varying graphics (and memory)
capacity.  Is there a way I can check the native resolution of the
machine
so that I can make adjustments to my code for the possible limitations of
the machine running it?

Only via C code, which is how R does it.
Hi,

Javascript knows, can we ask him ?

I mean, if I do that in R :

a - tempfile()
cat('htmlscript type=text/javascript document.write(screen.width)
;  /script/html', file=a)
browseURL(a)

I get 1920 in my browser's window. Can R read it ?

Romain

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Remove empty list from list - remove only one row and make as matrix

```try the following:

# a
dff[sapply(dff, is.matrix)]

# b
lapply(dff, function(x) if(!is.matrix(x)) rbind(x) else x)

I hope it helps.

Best,
Dimitris

Dimitris Rizopoulos
Ph.D. Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven

Tel: +32/(0)16/336899
Fax: +32/(0)16/337015
Web: http://med.kuleuven.be/biostat/
http://www.student.kuleuven.be/~m0390867/dimitris.htm

- Original Message -
To: r-help@stat.math.ethz.ch
Sent: Tuesday, August 29, 2006 1:22 PM
Subject: Re: [R] Remove empty list from list - remove only one row and
make as matrix

Dear all,
After I work around, I found in my list of data with only one row
which
need to remove or make it as matrix.
Here I write again my other toy example:
x - list(matrix(1:20, 5, 4),matrix(1:20, 5, 4),matrix(1:20, 5,
4),matrix(1:20, 5, 4),matrix(1:20, 5, 4))
y - list(c(1, -1, -1, 1, 1),c(1, -1, -1, -1, -1),c(1, 1, 1, 1,
1),c(1,
1, -1, 1, -1),c(-1, -1, -1, -1, -1))
## Thanks to Gabor Grothendieck for this trick.
## SIMPLIFY? SIMPLIFY  simplify
xy.list - mapply(cbind, x, y, SIMPLIFY=FALSE)

point.class -
t(cbind(c(10,20,15,4,-1),c(21,10,15,34,-1),c(11,13,6,3,1),c(7,5,5,2,1),c(8,9,5,12,-1)))
class.diffsame - points.neighb(as.matrix(point.class), xy.list, 5)
pd.class - points.diff(class.diffsame,xy.list)

nc.test - vector(list,length(pd.class))
for (i in 1:length(pd.class)) {
nc.test[[i]] - pd.class[[i]]\$point.diff
}
nc.test

# delete null/empty entries in a list
dff - delete.NULLs(nc.test)
dff; str(dff)
dff
[[1]]
[,1] [,2] [,3] [,4] [,5]
[1,]16   11   161
[2,]49   14   191
[3,]5   10   15   201

[[2]]
[1]  1  6 11 16  1

[[3]]
[,1] [,2] [,3] [,4] [,5]
[1,]38   13   18   -1
[2,]5   10   15   20   -1

lapply(dff, nrow)
lapply(dff, nrow)
[[1]]
[1] 3

[[2]]
NULL

[[3]]
[1] 2

#I can use
#dff[unlist(lapply(dff, nrow) == 1)] #2,3, etc

I have two questions here:
a. I need to remove dff[[2]]
b. How to make it as matrix (in list). I mean the result something
like

[[1]]
[,1] [,2] [,3] [,4] [,5]
[1,]16   11   161
[2,]49   14   191
[3,]5   10   15   201

[[2]]
[,1] [,2] [,3] [,4] [,5]
[1]1611   161

[[3]]
[,1] [,2] [,3] [,4] [,5]
[1,]38   13   18   -1
[2,]5   10   15   20   -1

points.neighb - function(p.class, list.nc, class.col) {
ntuples - nrow(p.class)
instvec - vector(list,length=ntuples)
for (i in 1:ntuples) {
# Thanks to Petr Pikal for this trick
instvec[[i]]\$class.diff - (p.class[i,class.col] -
list.nc[[i]][,class.col])!=0
instvec[[i]]\$class.same - (p.class[i,class.col] -
list.nc[[i]][,class.col])==0
}
instvec
}

points.diff - function(p.class, list.nc) {
ntuples - length(list.nc)
instvec - vector(list,ntuples)
for (i in 1:ntuples) {
instvec[[i]]\$point.diff -
list.nc[[i]][p.class[[i]]\$class.diff,]
instvec[[i]]\$point.same -
list.nc[[i]][p.class[[i]]\$class.same,]
}
instvec
}

# Thanks to Jim Holtman for this trick
delete.NULLs  -  function(x.list){
x.list[unlist(lapply(x.list, length) != 0)]
}

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] EOF and CCA analysis

```More than 10,000 locations but how many variables?

Like all computer programs there are limitations, but they are largely
imposed by your environment, so what OS and how much RAM do you have?

On Tue, 29 Aug 2006, isidora k wrote:

Hello!
I would like to do Empirical Orthogonal functions and
Canonical correlation analysis on satellite data. My
matrices are going to be very big (more than 10,000
locations). Are there limitations in R regarding the
size of the matrix?
thanks a lot
regards
Isidora

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

--
Brian D. Ripley,  [EMAIL PROTECTED]
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel:  +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax:  +44 1865 272595

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Remove empty list from list - remove only one row and make as matrix

```
Dear all,
Dimitris, thanks for your great help and quick response.

dff[sapply(dff, is.matrix)]
[[1]]
[,1] [,2] [,3] [,4] [,5]
[1,]16   11   161
[2,]49   14   191
[3,]5   10   15   201

[[2]]
[,1] [,2] [,3] [,4] [,5]
[1,]38   13   18   -1
[2,]5   10   15   20   -1

lapply(dff, function(x) if(!is.matrix(x)) rbind(x) else x)
[[1]]
[,1] [,2] [,3] [,4] [,5]
[1,]16   11   161
[2,]49   14   191
[3,]5   10   15   201

[[2]]
[,1] [,2] [,3] [,4] [,5]
x16   11   161

[[3]]
[,1] [,2] [,3] [,4] [,5]
[1,]38   13   18   -1
[2,]5   10   15   20   -1

On this day 29/08/2006 13:34, Dimitris Rizopoulos wrote:
try the following:

# a
dff[sapply(dff, is.matrix)]

# b
lapply(dff, function(x) if(!is.matrix(x)) rbind(x) else x)

I hope it helps.

Best,
Dimitris

Dimitris Rizopoulos
Ph.D. Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven

Tel: +32/(0)16/336899
Fax: +32/(0)16/337015
Web: http://med.kuleuven.be/biostat/
http://www.student.kuleuven.be/~m0390867/dimitris.htm

- Original Message -
To: r-help@stat.math.ethz.ch
Sent: Tuesday, August 29, 2006 1:22 PM
Subject: Re: [R] Remove empty list from list - remove only one row and
make as matrix

Dear all,
After I work around, I found in my list of data with only one row
which
need to remove or make it as matrix.
Here I write again my other toy example:
x - list(matrix(1:20, 5, 4),matrix(1:20, 5, 4),matrix(1:20, 5,
4),matrix(1:20, 5, 4),matrix(1:20, 5, 4))
y - list(c(1, -1, -1, 1, 1),c(1, -1, -1, -1, -1),c(1, 1, 1, 1,
1),c(1,
1, -1, 1, -1),c(-1, -1, -1, -1, -1))
## Thanks to Gabor Grothendieck for this trick.
## SIMPLIFY? SIMPLIFY  simplify
xy.list - mapply(cbind, x, y, SIMPLIFY=FALSE)

point.class -
t(cbind(c(10,20,15,4,-1),c(21,10,15,34,-1),c(11,13,6,3,1),c(7,5,5,2,1),c(8,9,5,12,-1)))
class.diffsame - points.neighb(as.matrix(point.class), xy.list, 5)
pd.class - points.diff(class.diffsame,xy.list)

nc.test - vector(list,length(pd.class))
for (i in 1:length(pd.class)) {
nc.test[[i]] - pd.class[[i]]\$point.diff
}
nc.test

# delete null/empty entries in a list
dff - delete.NULLs(nc.test)
dff; str(dff)
dff
[[1]]
[,1] [,2] [,3] [,4] [,5]
[1,]16   11   161
[2,]49   14   191
[3,]5   10   15   201

[[2]]
[1]  1  6 11 16  1

[[3]]
[,1] [,2] [,3] [,4] [,5]
[1,]38   13   18   -1
[2,]5   10   15   20   -1

lapply(dff, nrow)
lapply(dff, nrow)
[[1]]
[1] 3

[[2]]
NULL

[[3]]
[1] 2

#I can use
#dff[unlist(lapply(dff, nrow) == 1)] #2,3, etc

I have two questions here:
a. I need to remove dff[[2]]
b. How to make it as matrix (in list). I mean the result something
like

[[1]]
[,1] [,2] [,3] [,4] [,5]
[1,]16   11   161
[2,]49   14   191
[3,]5   10   15   201

[[2]]
[,1] [,2] [,3] [,4] [,5]
[1]1611   161

[[3]]
[,1] [,2] [,3] [,4] [,5]
[1,]38   13   18   -1
[2,]5   10   15   20   -1

points.neighb - function(p.class, list.nc, class.col) {
ntuples - nrow(p.class)
instvec - vector(list,length=ntuples)
for (i in 1:ntuples) {
# Thanks to Petr Pikal for this trick
instvec[[i]]\$class.diff - (p.class[i,class.col] -
list.nc[[i]][,class.col])!=0
instvec[[i]]\$class.same - (p.class[i,class.col] -
list.nc[[i]][,class.col])==0
}
instvec
}

points.diff - function(p.class, list.nc) {
ntuples - length(list.nc)
instvec - vector(list,ntuples)
for (i in 1:ntuples) {
instvec[[i]]\$point.diff -
list.nc[[i]][p.class[[i]]\$class.diff,]
instvec[[i]]\$point.same -
list.nc[[i]][p.class[[i]]\$class.same,]
}
instvec
}

# Thanks to Jim Holtman for this trick
delete.NULLs  -  function(x.list){
x.list[unlist(lapply(x.list, length) != 0)]
}

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] EOF and CCA analysis

```
I have got windows and 1Gb ram. my matrices are
monthly data for 10 years for sometimes more than
10,000 locations. I am going to perform the EOF on one
variable at a time and then the CCA on the principal
components that I will get from EOF. Do you think I
could do that in R?I could also find a macintosh I
could work with or get more RAM if you think this will
cause problems, but I am more concerned with the
software limitations if any!
thank you so much for your help.
Kind Regards
Isidora

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] MODWT exceeds sample size

```
Hello,

I get the following message when I try to load a time series of 128 stock
returns.

Error in modwt(stock, wf = la8, n.levels = 8, boundary = periodic) :
wavelet transform exceeds sample size in modwt

I thought that MODWT had no restrictions on the sample size?

Any help?

--
View this message in context:
http://www.nabble.com/MODWT-exceeds-sample-size-tf2183217.html#a6037934
Sent from the R help forum at Nabble.com.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] EOF and CCA analysis

```That's a 10,000 x 120 matrix, if I understand you right.  That is quite
modest.  The key issue is to be able to do an SVD:

A - matrix(rnorm(1e4*120), 1e4, 120)
dim(A)
[1] 1   120
system.time(svd(A))
[1] 2.79 0.13 2.93   NA   NA

on a similar machine to yours, using 50Mb.

Rblas.dll if you are doing much of this.

On Tue, 29 Aug 2006, isidora k wrote:

I have got windows and 1Gb ram. my matrices are
monthly data for 10 years for sometimes more than
10,000 locations. I am going to perform the EOF on one
variable at a time and then the CCA on the principal
components that I will get from EOF. Do you think I
could do that in R?I could also find a macintosh I
could work with or get more RAM if you think this will
cause problems, but I am more concerned with the
software limitations if any!
thank you so much for your help.
Kind Regards
Isidora

__
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com

--
Brian D. Ripley,  [EMAIL PROTECTED]
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel:  +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax:  +44 1865 272595

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] AffyChip Background Analysis

```Gunther Höning [EMAIL PROTECTED] writes:

Dear list,

I want to analyse some HG133Plus2.0 Affymetrix chips.
The first thing I intent to do, is just to perform a background correction
(mas, rma, gcrma,...) with the cel files.
Then I want to take a look at the files.
How can I do this ?

You might want to take a look at some of the packages in the
Bioconductor project.

Here is a link to the BioC packages that deal with visualization:
http://www.bioconductor.org/packages/release/Visualization.html

+ seth

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] First elements of a list.

```Suppose I have the following list:

a - strsplit(c(John;Smith, Jane;Doe, koda, gunner), ;)

I want to get to these two vectors without looping...

firstNames:c(John, Jane, koda, gunner)
lastNames:c(Jane, Doe, NA, NA)

Thanks

cn

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] First elements of a list.

```  sapply(a, [, 1)
[1] John   Jane   koda   gunner
sapply(a, [, 2)
[1] Smith Doe   NA  NA
---
Jacques VESLOT

CNRS UMR 8090
I.B.L (2ème étage)
1 rue du Professeur Calmette
B.P. 245
59019 Lille Cedex

Tel : 33 (0)3.20.87.10.44
Fax : 33 (0)3.20.87.10.31

http://www-good.ibl.fr
---

cory a écrit :
Suppose I have the following list:

a - strsplit(c(John;Smith, Jane;Doe, koda, gunner), ;)

I want to get to these two vectors without looping...

firstNames:c(John, Jane, koda, gunner)
lastNames:c(Jane, Doe, NA, NA)

Thanks

cn

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] symbols in coplots

```Kind folks,

I have a data frame of mean values for 14 levels and 5 factors. I'd like to
use a coplot to illustrate how the mean values change.

Can anyone provide an example of how I can specify different ploting symbols
in this plot.  I'd like to use either use a five different colors (col) for
the plotting symbol repeatedly in each of 14 panels, or 14 different symbols
(pch) for the different levels.  Either approach will work

Thank you.

--
Steve

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Bioconductor installation errors

```Hello,

some errors:

chmod: cannot access `/usr/local/lib/R/site-library/affyio/libs/*': No such
file or directory
ERROR: compilation failed for package 'affyio'
** Removing '/usr/local/lib/R/site-library/affyio'

* Installing *source* package 'survival' ...
** libs
ERROR: compilation failed for package 'survival'
** Removing '/usr/local/lib/R/site-library/survival'
* Installing *source* package 'KernSmooth' ...
** libs
ERROR: compilation failed for package 'KernSmooth'
** Removing '/usr/local/lib/R/site-library/KernSmooth'
* Installing *source* package 'RColorBrewer' ...
** R
** inst
** help
Building/Updating help pages for package 'RColorBrewer'
Formats: text html latex example
ColorBrewer   texthtmllatex   example
** building package indices ...
* DONE (RColorBrewer)
* Installing *source* package 'Biobase' ...
** libs
ERROR: compilation failed for package 'Biobase'
** Removing '/usr/local/lib/R/site-library/Biobase'
* Installing *source* package 'Biostrings' ...
** libs
ERROR: compilation failed for package 'Biostrings'
** Removing '/usr/local/lib/R/site-library/Biostrings'

..

..

Could you give me some advice about how to fix the errors?

Thanks.

Weigong Ge

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Bioconductor installation errors

```Ge, Weigong* [EMAIL PROTECTED] writes:

Hello,

some errors:

list:

http://www.bioconductor.org/docs/mailList.html

Your system (Linux?) is missing very basic development tools.  It
seems you have no make program installed.  Such a program, among many
others, is required to build R packages from source.  Try to install
system packages for software development (C complier, etc).

+ seth

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Bioconductor installation errors

```Hi,

You don't say, but the errors make me think you are using Linux.
you are missing some of the Essential programs under Unix -
they really are essential.

I think you need to reread the R admin manual, as instructed
and in particular this part:

Sarah

On 8/29/06, Ge, Weigong* [EMAIL PROTECTED] wrote:
Hello,

some errors:

--
Sarah Goslee
http://www.stringpage.com

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] spectral clustering

```Is there a function in R  that does spectral clustering?

Thanks ../Murli

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] spectral clustering

```Nair, Murlidharan T a écrit :
Is there a function in R  that does spectral clustering?

Thanks ../Murli

http://www.R-project.org/posting-guide.html **
Hi,

You would get a faster answer asking to RSiteSearch rather than this list.

Romain

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] passing namees

```R 2.3.1

I wrote a little script to do some cross correlations. The symbols are
in a text file like so:

symbols.txt
ibm
dd
csco

require(tseries)
symbols - scan(symbols.txt, what = 'character')

for(line in 1:(length(symbols)-1)) {
assign(symbols[line], get.hist.quote(instrument = symbols[line],
start = 2005-09-01, quote = Close))
}
# this results in objects ibm, dd... with the last symbol skipped

mat - cbind(symbols) # this is the problem
(cor_mat - cor(mat))
symnum(cor_mat)

How can I pass a list of the objects to cbind()? As written cbind gets
only the names of the objects and binds the names not the objects.

jab
--
John Bollinger, CFA, CMT
www.BollingerBands.com

If you advance far enough, you arrive at the beginning.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] how to contrast with factorial experiment

```Hello, R experts,
If I understand Ted's anwser correctly, then I can not contrast the
mean yields between sections 1-8 and 9-11 under Trt but I can
contrast mean yields for sections 1-3 and 6-11 because there exists
significant interaction between two factors  (Trt:section4,
Trt:section5). Could I use the commands below to test
the difference between sections 1-3 and 6-11 ?
contrasts(section)-c(-2,-2,-2,0,0,1,1,1,1,1,1)
newobj-lm(log2(yield)~treat*section)
How can I infer that there is significant difference between sections
1-3 and sections 6-11 for the Trt from the output below?

summary(newobj)

Call:
lm(formula = log2(yield) ~ treat * section)

Residuals:
Min   1Q   Median   3Q  Max
-0.49647 -0.14913 -0.01521  0.17471  0.51105

Coefficients:
Estimate Std. Error t value Pr(|t|)
(Intercept) 6.288400.05003 125.682   2e-16 ***
treatTrt1.221220.07076  17.259   2e-16 ***
section10.178310.03911   4.559 4.08e-05 ***
section2   -0.231020.16595  -1.392  0.17087
section32.381700.16595  14.352   2e-16 ***
section43.368340.16595  20.298   2e-16 ***
section5   -1.568730.16595  -9.453 3.67e-12 ***
section6   -0.415220.16595  -2.502  0.01613 *
section7   -0.899430.16595  -5.420 2.38e-06 ***
section80.095220.16595   0.574  0.56901
section9   -0.787840.16595  -4.748 2.21e-05 ***
section10   0.748210.16595   4.509 4.79e-05 ***
treatTrt:section1   0.101010.05532   1.826  0.07461 .
treatTrt:section2   0.272700.23468   1.162  0.25151
treatTrt:section3  -1.222100.23468  -5.207 4.85e-06 ***
treatTrt:section4  -1.391870.23468  -5.931 4.26e-07 ***
treatTrt:section5  -0.761370.23468  -3.244  0.00225 **
treatTrt:section6   0.073200.23468   0.312  0.75658
treatTrt:section7   0.331080.23468   1.411  0.16535
treatTrt:section8  -0.136860.23468  -0.583  0.56276
treatTrt:section9   0.220860.23468   0.941  0.35180
treatTrt:section10 -0.144760.23468  -0.617  0.54054
---
Signif. codes:  0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1

Residual standard error: 0.2874 on 44 degrees of freedom
Multiple R-Squared: 0.973,  Adjusted R-squared: 0.9601
F-statistic: 75.55 on 21 and 44 DF,  p-value:  2.2e-16

Joshua
Quoting (Ted Harding) [EMAIL PROTECTED]:

On 24-Aug-06 [EMAIL PROTECTED] wrote:
Hello, R users,
I have two factors (treat, section) anova design experiment where
there are 3 replicates. The objective of the experiment is to test if
there is significant difference of yield between top (section 9 to 11)
and bottom (section 9 to 11)
[I think you mean sections 1 to 8]

of the fruit tree under treatment. I found that there are interaction
between two factors. I wonder if I can contrast means from levels of
one factor (section) under another factor (treat)? if so, how to do
it in R and how to interpret the output?

I think you would be well advised to look at a plot of the data.
For example, let Y stand for yield, R for replicate, T for treat
and S for section.

ix-(T==Trt);plot(S[ix],Y[ix],col=red,ylim=c(0,1000))
ix-(T==Ctl);points(S[ix],Y[ix],col=blue)

From this it is clear that sections 4 and 5 are in a class of
their own. Also, in sections 1-3 and 6-11 the Ctl yields
are not only lower, but have smaller (in some cases hardly any)
variance, compared with the Trt yields. The variances for
sections 7,8,9,10,11 are greater than for 1,2,3,6 without
great change in mean value.

While there is an evident difference between Trt yields and
Ctrl yields for sections 1-3 and 6-11, this is not so for
sections 4 and 5.

This sort of behaviour no doubt provides some reasons for the
interaction you observed. You seem to have a quite complex
phenomenon here!

To some extent the problems with variance can be diminished by
working with logarithms. Compare the previous plot with

ix-(T==Trt);plot(S[ix],log10(Y[ix]),col=red,ylim=c(0,3))
ix-(T==Ctl);points(S[ix],log10(Y[ix]),col=blue)

(you have used log2() in your commands). The above observations
can be seen reflected in R if you look at the output of

summary(obj)

where in particular:

treatTrt:section2  -1.116910.33189  -3.365 0.001595 **
treatTrt:section3  -0.456340.33189  -1.375 0.176099
treatTrt:section4  -1.566270.33189  -4.719 2.42e-05 ***
treatTrt:section5  -1.736040.33189  -5.231 4.48e-06 ***
treatTrt:section6  -0.913110.33189  -2.751 0.008588 **
treatTrt:section7  -0.078530.33189  -0.237 0.814055
treatTrt:section8   0.179350.33189   0.540 0.591654
treatTrt:section9  -0.288590.33189  -0.870 0.389277
treatTrt:section10  0.069130.33189   0.208 0.835972
treatTrt:section11 -0.296490.33189  -0.893 0.376543

which, precisely, contrasts means from levels of one factor
(section) under another factor (treat), and shows that most
of the interaction arises ```

### [R] [R-pkgs] The rpanel package

```The rpanel package builds on the tcltk package to provide simple
interactive controls for R functions, in particular to provide simple
forms of dynamic graphics.  The intention is to make this form of
control particularly easy for R users to implement, with full
documentation.  The necessary tcltk variables are managed behind the
scenes so that users need not be concerned with any technicalities.

A descriptive paper and some simple example scripts are available at
The package is now available from CRAN.

Best wishes,

(on behalf of the authors: Bowman, Crawford, Alexander  Bowman)

___
Dept. of Statistics   Fax: +44-141-330-4814
The University of Glasgow E-mail:[EMAIL PROTECTED]
Glasgow G12 8QQ, U.K. Web: www.stats.gla.ac.uk/~adrian

___
R-packages mailing list
R-packages@stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-packages

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] passing namees

```Put the data into a list, not into individual variables.  Something
like this (untested):

L - sapply(symbols[-length(symbols)],
get.hist.quote, start = 2005-01-09, quote = Close, simplify = FALSE)
mat - do.call(cbind, L)

cor.mat - cor(mat, use = complete)
symnum(cor.mat)

On 8/29/06, BBands [EMAIL PROTECTED] wrote:
R 2.3.1

I wrote a little script to do some cross correlations. The symbols are
in a text file like so:

symbols.txt
ibm
dd
csco

require(tseries)
symbols - scan(symbols.txt, what = 'character')

for(line in 1:(length(symbols)-1)) {
assign(symbols[line], get.hist.quote(instrument = symbols[line],
start = 2005-09-01, quote = Close))
}
# this results in objects ibm, dd... with the last symbol skipped

mat - cbind(symbols) # this is the problem
(cor_mat - cor(mat))
symnum(cor_mat)

How can I pass a list of the objects to cbind()? As written cbind gets
only the names of the objects and binds the names not the objects.

jab
--
John Bollinger, CFA, CMT
www.BollingerBands.com

If you advance far enough, you arrive at the beginning.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Write signed short into a binary file (follow up and conclusion)(for real)

```Somebody pointed me to the right spot in the documentation, and kindly and
humorously
'suggested' to me to do the following:
My Apologies, everybody!
I should have read docs a bit better, actually...
...will do better next time!
luca

- Original Message -
From: [EMAIL PROTECTED]
To:
Sent: Monday, August 28, 2006 4:52 PM
Subject: Write signed short into a binary file (follow up and conclusion)

I've solved my problem using:
(purpose: write the signed integer -19 as a two byte integer into a binary
file)

writeBin(as.integer(-19),myconnection, size=2)

There is no need to coerce.

Thanks anyway!

Luca NanettiUniversity Medical Center Groningen
BCN-NeuroImagingCenter
A.Deusinglaan 2 9713AW Groningen
The Nethterlands
l dot nanetti at med dot rug dot nl

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] forestplot fucntion in rmeta package

```Dear R users,

I would like to adjust the x axis the way I wanted using
forestplot(labeltext, mean, lower, upper, align = NULL, xlab = , zero =
0, graphwidth = unit(2, inches), col = meta.colors(), xlog = FALSE). I
tried using xaxt=n and then redefine the axis using axis(1, at=c(0,0.5,1.0
,1.5,2.0,2.5),label=c(0,0.5,1.0,1.5,2.0,2.5)), but it doesn't work at all.
Seems the forestplot function itself does not allow some basic plot option
like xaxt=n? Anybody porvides some hints to solve this problem? Many
thanks!

Shirley

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Dendrogram troubles

```Hi,
I am making a dendrogram with 180 terminal values. Whether I keep it
horizontal or vertical, it gives a 'squished' graph that refuses to be
stretched beyond the window size and I cannot read the labels.
(I'm using hclust and then plot to make the tree.)
Is there a way to stretch the graph in R, or by exporting it somehow, so
that I can read the 180 values and see what the results are?
Thank you very much.
-DS.

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Key() and par(mfrow)

```Hi Folks,

I want to use key() to position the legend on top of the page in the following
example. Prefer not to use locator() par(mfrow = c(2,2)) plot(1:4,
title=Plot Number1) plot(rnorm(10), title=Plot Number2)  plot(1:10,
title=Plot Number3)  key(x,y, text=list(c(Alpha,Beta)),
text=list(c(1,2))) x,y should indicate the position of the legend which
is top left of the   page. The lists here are oversimplified but we have to
use key(). Thank you for your help!
_
Search from any Web page with powerful protection. Get the FREE Windows Live
Toolbar Today!

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] lattice/xyplot: plotting 4 variables in two panels - can this be done?

```Hi,
I would like to create a plot of y1,y2,y3,y4 against x for several subjects
such that y1 and y2 are plotted against x in one panel and y3 and y4 against x
in another panel. Thus if there are 3 subjects I should end up with 6 panels.
Is there a simple way of doing so (i.e. without calling xyplot() several times,
and then padding the results together)??
Regards
Søren

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] subset by two variables

```I'm using this syntax to create data subsets for plots:
subset=source==Both. Now I would like to create a subset defined by
two different variables like this: subset=source==Both
subset=site==home but this syntax is not correct. The documents in the
manual for subset seem to be creating whole new data files not just
selecting rows based on the contents of the variables. How do I subset
based on two variables.

--
Dean Sonneborn, MS
Programmer Analyst
Department of Public Health Sciences
University of California, Davis
(530) 754-9516

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] lattice and several groups

```Dear R-list,

I would like to use the lattice library to show several groups on
the same graph. Here's my example :

## the data
f1 - factor(c(mod1,mod2,mod3),levels=c(mod1,mod2,mod3))
f1 - rep(f1,3)
f2 - factor(rep(c(g1,g2,g3),each=3),levels=c(g1,g2,g3))
df - data.frame(val=c(4,3,2,5,4,3,6,5,4), x=rep(c(1,2,3),3),f1=f1,f2=f2)
#
library(lattice)

para.liste - trellis.par.get()
superpose.symbol - para.liste\$superpose.symbol
superpose.symbol\$pch - c(1,2,3)
trellis.par.set(superpose.symbol,superpose.symbol)

# Now I can see the group according to the f1 factor (with a different
symbol for every modality)
xyplot( val~x,
data=df,
group=f1,
auto.key=list(space=right)
)

# or I can see the group according to the f2 factor
xyplot( val~x,
data=df,
type=l,
group=f2,
auto.key=list(space=right,points=FALSE,lines=TRUE)
)

How can I do to highlight both the f1 and f2 factors on one panel with
the legends, using the lattice function ?

Thanks

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Producing R demos

```Hello!

I have found terrific demo or R package functionality at

Author has told me off-list that he is using SnapzProX (on mac), and
ghostwriter (http://had.co.nz/ghostwriter/) to automate the typing.
Unfortunatelly, I do not have mac ;) Can anyone on the list suggest, how
such a demo (video + automated typing of a script) could be produced
under either Windows or Linux OS? I would like to create such a demo for
presentation as I will not be able to install R on the machine, but I
could play a movie.

Thanks!

--
Lep pozdrav / With regards,
Gregor Gorjanc

--
University of Ljubljana PhD student
Biotechnical Faculty
Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan
Groblje 3   mail: gregor.gorjanc at bfro.uni-lj.si

SI-1230 Domzale tel: +386 (0)1 72 17 861
Slovenia, Europefax: +386 (0)1 72 17 888

--
One must learn by doing the thing; for though you think you know it,
you have no certainty until you try. Sophocles ~ 450 B.C.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] my email: jonathan gheyssens (université de Montreal)

```
Jonathan Gheyssens
M.Sc Economie Appliquée
IEA - HEC Montréal
[EMAIL PROTECTED]
514-581-8009
514-678-5231

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] lattice and several groups

```  I would like to use the lattice library to show several groups on
the same graph. Here's my example :

## the data
f1 - factor(c(mod1,mod2,mod3),levels=c(mod1,mod2,mod3))
f1 - rep(f1,3)
f2 - factor(rep(c(g1,g2,g3),each=3),levels=c(g1,g2,g3))
df - data.frame(val=c(4,3,2,5,4,3,6,5,4), x=rep(c(1,2,3),3),f1=f1,f2=f2)

It's pretty easy to do this with ggplot:

install.packages(ggplot, repos=http://ggobi.org/r/;)
library(ggplot)
qplot(x, val, data=df, shape=f2, colour=f1)

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Substring and strsplit

```Dear R People:

I am trying to split a character vector into a set of individual
letters:

Ideal:
x3 - c(dog)
d o g

I tried the following:
strsplit(x3)
Error in strsplit(x3) : argument split is missing, with no default
strsplit(x3,1)
[[1]]
[1] dog

I know that this is incredibly simple, but what am I doing wrong?

Either Windows or Linux 2.3.1

Sincerely,
Erin Hodgess
Associate Professor
Department of Computer and Mathematical Sciences
University of Houston - Downtown
mailto: [EMAIL PROTECTED]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Substring and strsplit

```On Tue, 29 Aug 2006, Erin Hodgess wrote:

Dear R People:

I am trying to split a character vector into a set of individual
letters:

Ideal:
x3 - c(dog)
d o g

I tried the following:
strsplit(x3)
Error in strsplit(x3) : argument split is missing, with no default
strsplit(x3,1)
[[1]]
[1] dog

I know that this is incredibly simple, but what am I doing wrong?

This is the first example on the help page for strsplit.

-thomas

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Substring and strsplit

```Use '' as parameter to strsplit

x3 - 'dog'
strsplit(x3, '')
[[1]]
[1] d o g

On 8/29/06, Erin Hodgess [EMAIL PROTECTED] wrote:
Dear R People:

I am trying to split a character vector into a set of individual
letters:

Ideal:
x3 - c(dog)
d o g

I tried the following:
strsplit(x3)
Error in strsplit(x3) : argument split is missing, with no default
strsplit(x3,1)
[[1]]
[1] dog

I know that this is incredibly simple, but what am I doing wrong?

Either Windows or Linux 2.3.1

Sincerely,
Erin Hodgess
Associate Professor
Department of Computer and Mathematical Sciences
University of Houston - Downtown
mailto: [EMAIL PROTECTED]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

--
Jim Holtman
Cincinnati, OH
+1 513 646 9390

What is the problem you are trying to solve?

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] lattice/xyplot: plotting 4 variables in two panels - can this be done?

```On 8/29/06, Søren Højsgaard [EMAIL PROTECTED] wrote:
Hi,
I would like to create a plot of y1,y2,y3,y4 against x for several subjects
such that y1 and y2 are plotted against x in one panel and y3 and y4 against
x in another panel. Thus if there are 3 subjects I should end up with 6
panels. Is there a simple way of doing so (i.e. without calling xyplot()
several times, and then padding the results together)??

You need to transform your data into a suitable form. See ?reshape and
?make.groups.

Deepayan

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Handling realisations in geoRglm

```
Dear R users:

I want to model mosquito count data based on landcover attributes and
meteorological variables using a Poisson GLSM in the geoRglm package.  I
have monthly mosquito counts over more than 20 years with repeated
observations from individual trap sites over time.  I have used
as.geodata() to successfully read my dataset into the geodata format
utilized by geoR and geoRglm, including specification of the
realisations.  (I know the realisations have been specified correctly
because I can get data summaries by realisation from the geodata
object.)  However, after reading the packages' documentation and
searching the mailing lists and other sources, it seems that the
functions in geoRglm do not acknowledge the existence of multiple
realisations per site.  I see that the likfit() function in geoR has an
argument for realisations, but I cannot find anything similar in
geoRglm.  Is it possible to model data with repeated realisations from
the same sites using geoRglm?  If so, how can this be done and is there
a way to model dependence among the realisations over time?

Thank you,
Chris Barker
Ph.D. Candidate
University of California, Davis
[EMAIL PROTECTED]
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Help on apply() function

```Respected Sir/Madam,

I have a problem with apply function. I have to two matrices of dimension of
one column but n rows. I have to check whether one matrix is greater than
other by going thru each row (ie) using if condition to check one matrix
with another matrix.

I like to use apply() function to this approach. That is apply function
between two matrices. I searched for examples online but I couldn't find
any.

I don't know how to loop thru the matrices.

Any help is appreciated.

My mailid is [EMAIL PROTECTED]

Thanks,
Anusha.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] lattice and several groups

```Try this:

xyplot(val ~ x, data = df, type = p,
col = as.numeric(df\$f1), pch = as.numeric(df\$f2))

key1 - list(border = TRUE, colums = 2, text = list(levels(df\$f1)),
points = list(pch = 1:nlevels(df\$f1))
)

key2 - list(border = TRUE, colums = 2, text = list(levels(df\$f2)),
points = list(pch = 20, col = 1:nlevels(df\$f2))
)

trellis.focus(panel, 1, 1)
draw.key(key1, draw = TRUE, vp = viewport(.9, .9))
draw.key(key2, draw = TRUE, vp = viewport(.75, .9))
trellis.unfocus()

On 8/29/06, Laurent Rhelp [EMAIL PROTECTED] wrote:
Dear R-list,

I would like to use the lattice library to show several groups on
the same graph. Here's my example :

## the data
f1 - factor(c(mod1,mod2,mod3),levels=c(mod1,mod2,mod3))
f1 - rep(f1,3)
f2 - factor(rep(c(g1,g2,g3),each=3),levels=c(g1,g2,g3))
df - data.frame(val=c(4,3,2,5,4,3,6,5,4), x=rep(c(1,2,3),3),f1=f1,f2=f2)
#
library(lattice)

para.liste - trellis.par.get()
superpose.symbol - para.liste\$superpose.symbol
superpose.symbol\$pch - c(1,2,3)
trellis.par.set(superpose.symbol,superpose.symbol)

# Now I can see the group according to the f1 factor (with a different
symbol for every modality)
xyplot( val~x,
data=df,
group=f1,
auto.key=list(space=right)
)

# or I can see the group according to the f2 factor
xyplot( val~x,
data=df,
type=l,
group=f2,
auto.key=list(space=right,points=FALSE,lines=TRUE)
)

How can I do to highlight both the f1 and f2 factors on one panel with
the legends, using the lattice function ?

Thanks

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Producing R demos

```I have not used this but wink is a free demo-maker for Windows
available at:
http://www.debugmode.com/wink/
Click the red on green arrow on that page to a (very short) demo.

On 8/29/06, Gregor Gorjanc [EMAIL PROTECTED] wrote:
Hello!

I have found terrific demo or R package functionality at

Author has told me off-list that he is using SnapzProX (on mac), and
ghostwriter (http://had.co.nz/ghostwriter/) to automate the typing.
Unfortunatelly, I do not have mac ;) Can anyone on the list suggest, how
such a demo (video + automated typing of a script) could be produced
under either Windows or Linux OS? I would like to create such a demo for
presentation as I will not be able to install R on the machine, but I
could play a movie.

Thanks!

--
Lep pozdrav / With regards,
Gregor Gorjanc

--
University of Ljubljana PhD student
Biotechnical Faculty
Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan
Groblje 3   mail: gregor.gorjanc at bfro.uni-lj.si

SI-1230 Domzale tel: +386 (0)1 72 17 861
Slovenia, Europefax: +386 (0)1 72 17 888

--
One must learn by doing the thing; for though you think you know it,
you have no certainty until you try. Sophocles ~ 450 B.C.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Producing R demos

```Gabor Grothendieck wrote:
I have not used this but wink is a free demo-maker for Windows
available at:
http://www.debugmode.com/wink/
Click the red on green arrow on that page to a (very short) demo.

Thanks! Do you perhaps also know equivalent of ghostwriter (link is
bellow) for windows/linux?

On 8/29/06, Gregor Gorjanc [EMAIL PROTECTED] wrote:
Hello!

I have found terrific demo or R package functionality at

Author has told me off-list that he is using SnapzProX (on mac), and
ghostwriter (http://had.co.nz/ghostwriter/) to automate the typing.
Unfortunatelly, I do not have mac ;) Can anyone on the list suggest, how
such a demo (video + automated typing of a script) could be produced
under either Windows or Linux OS? I would like to create such a demo for
presentation as I will not be able to install R on the machine, but I
could play a movie.

Thanks!

--
Lep pozdrav / With regards,
Gregor Gorjanc

--
University of Ljubljana PhD student
Biotechnical Faculty
Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan
Groblje 3   mail: gregor.gorjanc at bfro.uni-lj.si

SI-1230 Domzale tel: +386 (0)1 72 17 861
Slovenia, Europefax: +386 (0)1 72 17 888

--
One must learn by doing the thing; for though you think you know it,
you have no certainty until you try. Sophocles ~ 450 B.C.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] C compile problem on Ubuntu linux

```
On 20 August 2006 at 15:29, Frank E Harrell Jr wrote:
| Manuel Castejón Limas wrote:
|  Hello,
|  I've just compiled Hmisc ok under dapper.
|  I think you need to further install some packages.
|  Have you installed libc6-dev?
|  I would start installing the build-essential package.
|  Best wishes
|  Manuel
|
| Thanks Manuel, apt-get install build-essential solved the problem.

FWIW r-base-dev also depends on build-essential, so problem-solving tip
number one (if on Debian and building things for R, install r-base-dev)
still applies.

Cheers, Dirk

--
Hell, there are no rules here - we're trying to accomplish something.
-- Thomas A. Edison

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Producing R demos

```Don't know.  I would check whether the functionality is included in wink.

On 8/29/06, Gregor Gorjanc [EMAIL PROTECTED] wrote:
Gabor Grothendieck wrote:
I have not used this but wink is a free demo-maker for Windows
available at:
http://www.debugmode.com/wink/
Click the red on green arrow on that page to a (very short) demo.

Thanks! Do you perhaps also know equivalent of ghostwriter (link is
bellow) for windows/linux?

On 8/29/06, Gregor Gorjanc [EMAIL PROTECTED] wrote:
Hello!

I have found terrific demo or R package functionality at

Author has told me off-list that he is using SnapzProX (on mac), and
ghostwriter (http://had.co.nz/ghostwriter/) to automate the typing.
Unfortunatelly, I do not have mac ;) Can anyone on the list suggest, how
such a demo (video + automated typing of a script) could be produced
under either Windows or Linux OS? I would like to create such a demo for
presentation as I will not be able to install R on the machine, but I
could play a movie.

Thanks!

--
Lep pozdrav / With regards,
Gregor Gorjanc

--
University of Ljubljana PhD student
Biotechnical Faculty
Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan
Groblje 3   mail: gregor.gorjanc at bfro.uni-lj.si

SI-1230 Domzale tel: +386 (0)1 72 17 861
Slovenia, Europefax: +386 (0)1 72 17 888

--
One must learn by doing the thing; for though you think you know it,
you have no certainty until you try. Sophocles ~ 450 B.C.
--

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] problem in install on ubuntu

```
On 24 August 2006 at 11:35, Ryan Austin wrote:
| Hi Bruno,
|
| Your missing the Basic Linear Algebra Subroutines 3.0 package.
|
| apt-get install refblas3
|
| should fix the problem. (At least in Debian, should be the same for Ubuntu)

The 'empty' package r-base-dev depends on 'refblas3-dev | atlas3-base-dev' so
installing r-base-dev on Debian or Ubuntu provides what most if not all
packages require.

Hth, Dirk

--
Hell, there are no rules here - we're trying to accomplish something.
-- Thomas A. Edison

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] lattice/xyplot: plotting 4 variables in two panels - can this be done?

```Try xyplot.zoo in the zoo package (be sure you have
the latest zoo version). One caveat is that xyplot.zoo is
under development and could change.  Here we use the
builtin data set anscombe as an example:

# display the anscombe data set
anscombe

library(lattice)
library(zoo)
# create zoo object using column 1 as x/times and columns 5:8 as the y's
z - zoo(data.matrix(anscombe[,5:8]), anscombe[,1])
xyplot(z, screens = c(1, 1, 2, 2), col = 1:4, pch = 1:4, type = b)

On 8/29/06, Søren Højsgaard [EMAIL PROTECTED] wrote:
Hi,
I would like to create a plot of y1,y2,y3,y4 against x for several subjects
such that y1 and y2 are plotted against x in one panel and y3 and y4 against
x in another panel. Thus if there are 3 subjects I should end up with 6
panels. Is there a simple way of doing so (i.e. without calling xyplot()
several times, and then padding the results together)??
Regards
Søren

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] lattice and several groups

```Note that before entering this you need:

library(lattice)
library(grid) # to access the viewport function

On 8/29/06, Gabor Grothendieck [EMAIL PROTECTED] wrote:
Try this:

xyplot(val ~ x, data = df, type = p,
col = as.numeric(df\$f1), pch = as.numeric(df\$f2))

key1 - list(border = TRUE, colums = 2, text = list(levels(df\$f1)),
points = list(pch = 1:nlevels(df\$f1))
)

key2 - list(border = TRUE, colums = 2, text = list(levels(df\$f2)),
points = list(pch = 20, col = 1:nlevels(df\$f2))
)

trellis.focus(panel, 1, 1)
draw.key(key1, draw = TRUE, vp = viewport(.9, .9))
draw.key(key2, draw = TRUE, vp = viewport(.75, .9))
trellis.unfocus()

On 8/29/06, Laurent Rhelp [EMAIL PROTECTED] wrote:
Dear R-list,

I would like to use the lattice library to show several groups on
the same graph. Here's my example :

## the data
f1 - factor(c(mod1,mod2,mod3),levels=c(mod1,mod2,mod3))
f1 - rep(f1,3)
f2 - factor(rep(c(g1,g2,g3),each=3),levels=c(g1,g2,g3))
df - data.frame(val=c(4,3,2,5,4,3,6,5,4), x=rep(c(1,2,3),3),f1=f1,f2=f2)
#
library(lattice)

para.liste - trellis.par.get()
superpose.symbol - para.liste\$superpose.symbol
superpose.symbol\$pch - c(1,2,3)
trellis.par.set(superpose.symbol,superpose.symbol)

# Now I can see the group according to the f1 factor (with a different
symbol for every modality)
xyplot( val~x,
data=df,
group=f1,
auto.key=list(space=right)
)

# or I can see the group according to the f2 factor
xyplot( val~x,
data=df,
type=l,
group=f2,
auto.key=list(space=right,points=FALSE,lines=TRUE)
)

How can I do to highlight both the f1 and f2 factors on one panel with
the legends, using the lattice function ?

Thanks

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Problem plotting with xyplot

```This is in regard to Dan Norlund's post from December 2003 (available at
https://stat.ethz.ch/pipermail/r-help/attachments/20031225/1bce02b8/attachment.pl

)

While I agree with Prof. Ripley's warning against fitting so few
observations per panel, this is not actually what causes the error.
Instead, the culprit is the fact that panel.loess uses
family='symmetric' by default; since the first iteration of fitting
interpolates the data, the residuals are zero, hence the robustness
weights are not defined for the next iteration. In the example Dan gave,
a change to

xyplot(tolerance~age | factor(id), data=tolerance.pp,
type=c(p,smooth), family=gaussian)

avoids this. As family='gaussian' is the default outside of panel.loess,
it is easy to forget this detail.

Ben

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### Re: [R] Help on apply() function

```Anusha anu1978 at hotmail.com writes:

I have a problem with apply function. I have to two matrices of dimension of
one column but n rows. I have to check whether one matrix is greater than
other by going thru each row (ie) using if condition to check one matrix
with another matrix.

I like to use apply() function to this approach. That is apply function
between two matrices. I searched for examples online but I couldn't find
any.

I don't know how to loop thru the matrices.

Any help is appreciated.

My mailid is monkponu at yahoo.com.

Thanks,
Anusha.

You can use the functions all.equal with the function isTRUE (see ?all.equal) to
check if two objects are nearly equal (within a certain tolerance). Or you can
use identical (see ?identical) to check if they are exactly the same. See the
examples in the help for identical.

Mark Lyman

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Need help to estimate the Coef matrices in mAr

```Dear R users,

I am using mAr package to fit a Vector autoregressive model to my data. But
here I want to put some predetermined values for some elements in
coefficient matrix that mAr.est going to estimate. For example if p=3 then I
want to put A3[1,3] = 0 and keep rest of the elements of coefficient
matrices to be determined by mAr.est.

Can anyone please tell me how can I do that?

Sincerely yours,
Arun

[[alternative HTML version deleted]]

__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

```

### [R] Cross-correlation between two time series data

```
Hi all,

I  have  two  time  series  data  (say  x  and  y). I am interested to
calculate the correlation between them and its confidence interval (or
to  test  no  correlation). Function cor.test(x,y) does the test of no
correlation. But this test probably is wrong because of autocorrelated
data.

ccf()  calculates the correlation between two series data. But it does
not  provide  the  confidence intervals of cross correlation. Is there
any  function  that  calculates the confidence interval of correlation
between  two  time  series data or performs the test of no correlation
between two time series data.

Thanks.

Jun
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help