Hello,
Is there an iterative reweighting option in LOESS function from modreg
package to perform robust fitting in the presence of outliers in the data ?
Thanks in advance.
Philippe.
__
[EMAIL PROTECTED] mailing list
On Fri, 14 Mar 2003, Philippe Hupé wrote:
Is there an iterative reweighting option in LOESS function from modreg
package to perform robust fitting in the presence of outliers in the data ?
See the help page, the reference on the help page and the source code.
I guess the answer is `not quite
Dear all,
I fit independent GLMs for a 2x2 factorial problem on the data matrix of
size 9500 x 12 (genes x arrays) and get 9500 observed t-values using the
apply() function. Now, I wish to get the permutated p-values. Therefore
I random sample the class labels and perform the glm fitting to get
On Fri, 14 Mar 2003, Philippe Hupé wrote:
I have just found the answer : iterative reweighting can be set through
loess.control function via iterations option.
No so: it's through the family argument to loess: `iterations' sets how
many iterations *when* robust fitting is selected.
This is
The problem in this case is that the RPM file is corrupted. Try
downloading a fresh copy from CRAN.
To check if the package has been corrupted type:
rpm -K --nosignature /mnt/cdrom/linux/8.x/R-1.6.2-1.i386.rpm
It is possible to install a gpg-signed RPM package without importing the
On Fri, 14 Mar 2003, wolski wrote:
I am trying to plot two graphics with different sizes to one devise.
par(mfrow=c(1,2))
# is set the size for the first graphic
par(fig=c(0,7,0,10)/10)
#and plot it
plot(1,1)
#then i set the size
par(fig=c(7,10,0,10)/10)
#and plot the second one
Dear all,
I've seen that there are some maps, or at least costlines in R. The oz
package is the map of Autralia and in the fields package there is a US
map and a world map. This world map allows to select smaller sections
such Europe:
library(fields)
world(xlim=c(-10,18),ylim=c(36,60))
You can try RarcInfo, which has it's homepage at
http://matheron.estadi.uv.es/~virgil/Rpackages/RArcInfo/index.shtml.
there is a link to free data in the section related resources.
regards
Mathieu Roelants
Mathieu Roelants - Project Vlaamse Groeicurven
Laboratory of Anthropogentics
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Martyn Plummer
Sent: Friday, March 14, 2003 4:33 AM
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]; Wenyan Zhong
Subject: Re: [R] error installing R on linux 8.0
The problem in this case is that the RPM file
Hello R user,
I downloaded grasper_0.3-2.tar.gz and installed it as described.
$ su
$ R CMD INSTALL grasper_0.3-2.tar.gz
$ R
that seems to work, but
grasp.GUI()
doesn't work. I read the note, that graspeR need mgcv (downloaded),
MASS, modreg and tcktk . I haven't found the last 3 packetages in
Martin Wegmann wrote:
Hello R user,
I downloaded grasper_0.3-2.tar.gz and installed it as described.
$ su
$ R CMD INSTALL grasper_0.3-2.tar.gz
$ R
that seems to work, but
grasp.GUI()
doesn't work. I read the note, that graspeR need mgcv (downloaded),
MASS, modreg and tcktk . I haven't found the
MASS is part of VR, and all of VR, mgcv, modreg and tcltk (sic) are
part of all of every R installation (or should be).
However, tcltk will not work unless you have Tcl/Tk installed.
On Fri, 14 Mar 2003, Martin Wegmann wrote:
Hello R user,
I downloaded grasper_0.3-2.tar.gz and installed it
Hello
1*
I need to use logistic regression. But
my data file is very huge( appx. 4 million line).
R doesn't handle such a file.
What can I do ?
2*
So, I thought whether I could perform sta. analyses on summarised
data (count of yes/no values) of the huge file. Normally,
Hello,
I have Tcl/tk installed but it is version 8.4. I already have a problem
with NVIZ in GRASS which requires 8.3 and doesn't work now. Same problem
with grasp?
doesn't work means that R responds after I typed in
grasp.GUI()
command not recognized (or similar prompt)
thanks martin
[EMAIL
Hi.
We've written a GUI for R. Under linux and Macintosh, it looks good, but
under Windows, the rendering is unbearably slow.
Can we do anything to speed it up?
Is it possible to compile our program, for instance?
If not, I have a second question:
The buttons on our GUI move over
On Fri, 14 Mar 2003, orkun wrote:
1*
I need to use logistic regression. But
my data file is very huge( appx. 4 million line).
R doesn't handle such a file.
What can I do ?
R does handle such files (which are tiny by data-mining standards): you
just need to put 1GB or 2GB of memory in your
I have a question for our European readers: how common is it to use
commas as decimal points in spread-sheet and statistics applications?
Is it an inconvenience to require that all data use a period as decimal point?
(i.e., 3.14159 rather than 3,14159).
We're trying to make our program as
I'm having a weird problem with length(), in R1.6.1 under windows2000. I have a
dataframe called byyr, with ten columns, the first of which is named cnd95.
summary(byyr) shows that byyr$cnd95 contains the factor level tr 66 times. Also,
when I enter byyr$cnd95 at the command line, I can count 66
A little while ago, I asked for help on producing biplots with more
options than are provided by the function biplot in library mva. No help
was forthcoming, so I wrote my own function, essentially a lattice version
of biplot with some extra features for plotting axes. In case anyone is
On Fri, 14 Mar 2003, janet rosenbaum wrote:
I have a question for our European readers: how common is it to use
commas as decimal points in spread-sheet and statistics applications?
Is it an inconvenience to require that all data use a period as decimal point?
(i.e., 3.14159 rather than
It's the users who are misbehaving -- it usually is!
I think you mean [byyr$cnd95 %in% tr], not the same thing as R has
NA character strings.
x - c(a, a, NA, b2)
x == a
[1] TRUE TRUENA FALSE
x[x == a]
[1] a a NA
x[x %in% a]
[1] a a
MASS4 page 30 discusses this and similar traps.
On
David Parkhurst wrote:
I'm having a weird problem with length(), in R1.6.1 under windows2000. I have a
dataframe called byyr, with ten columns, the first of which is named cnd95.
summary(byyr) shows that byyr$cnd95 contains the factor level tr 66 times. Also,
when I enter byyr$cnd95 at the
David Parkhurst wrote:
With the problem below, I've discoved that
n95trt-length(byyr$cnd95[byyr$cnd95==tr!is.na(byyr$cnd95)])
does give me the correct count for the number of tr entries. (The same behavior
occurs for the c level of the cnd95 factor.) It appears that
byyr$cnd95==tr
is finding
HI Dave,
| From: David Parkhurst [EMAIL PROTECTED]
| Date: Fri, 14 Mar 2003 10:35:19 -0500
|
| I'm having a weird problem with length(), in R1.6.1 under windows2000. I have a
| dataframe called byyr, with ten columns, the first of which is named cnd95.
| summary(byyr) shows that
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of David
Parkhurst
Sent: Friday, March 14, 2003 9:35 AM
To: [EMAIL PROTECTED]
Subject: [R] length() misbehaving?
I'm having a weird problem with length(), in R1.6.1 under
windows2000. I have a dataframe
-Original Message-
From: Marc Schwartz [mailto:[EMAIL PROTECTED]
Sent: Friday, March 14, 2003 10:23 AM
To: 'David Parkhurst'; '[EMAIL PROTECTED]'
Subject: RE: [R] length() misbehaving?
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of David
On Fri, 14 Mar 2003 16:48:37 +0200
orkun [EMAIL PROTECTED] wrote:
Hello
1*
I need to use logistic regression. But
my data file is very huge( appx. 4 million line).
R doesn't handle such a file.
What can I do ?
2*
So, I thought whether I could perform sta.
[EMAIL PROTECTED] writes:
On Fri, 14 Mar 2003, janet rosenbaum wrote:
I have a question for our European readers: how common is it to use
commas as decimal points in spread-sheet and statistics applications?
Is it an inconvenience to require that all data use a period as decimal
On Fri, 14 Mar 2003, David Parkhurst wrote:
I'm having a weird problem with length(), in R1.6.1 under windows2000. I have a
dataframe called byyr, with ten columns, the first of which is named cnd95.
summary(byyr) shows that byyr$cnd95 contains the factor level tr 66 times. Also,
when I
Frank E Harrell Jr [EMAIL PROTECTED] writes:
If you have no more than one continuous variable you can pre-process
(outside of R) to collapse the data into frequency counts. I did not
check to see if glm handles frequency case weights. The lrm function
in the Design package
In a message dated 3/14/03 10:10:15 AM Eastern Standard Time,
[EMAIL PROTECTED] writes:
1*
I need to use logistic regression. But
my data file is very huge( appx. 4 million line).
R doesn't handle such a file.
What can I do ?
It depends on the strength of your computing system as well:
Thanks for all the helpful decimal point information. I implemented it,
so our friends at WHO should be happy.
It wasn't pure laziness --- I wanted to avoid cluttering the GUI.
To paraphrase from Ethics of the Fathers, the more features; the more worry.
Thanks,
Janet
On 14 Mar 2003, Peter Dalgaard BSA wrote:
Frank E Harrell Jr [EMAIL PROTECTED] writes:
If you have no more than one continuous variable you can pre-process
(outside of R) to collapse the data into frequency counts. I did not
check to see if glm handles frequency case weights. The lrm
On Fri, 14 Mar 2003, Edmond Ng wrote:
Hi all,
I have written a program which performs some data simulation, model fitting (to the
simulated data) and then it will save the parameter estimates from each loop into a
matrix for later use. Because convergence will not be met in some sets of
From ?on.exit:
`on.exit' records the expression given as its argument as needing
to be executed when the current function exits (either naturally
or as the result of an error).
Thus, when i - i+1, it then goes to while, which then exits,
invoking the on.exit command as it does so.
Thomas Lumley [EMAIL PROTECTED] writes:
On 14 Mar 2003, Peter Dalgaard BSA wrote:
Frank E Harrell Jr [EMAIL PROTECTED] writes:
If you have no more than one continuous variable you can pre-process
(outside of R) to collapse the data into frequency counts. I did not
check to see
Is it possible to display multiple plots in different windows rather than using the
split.screen ?
thanks.
-Geetha.
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
On Fri, 14 Mar 2003, Vasudevan, Geetha wrote:
Is it possible to display multiple plots in different windows rather than using the
split.screen ?
thanks.
Yes. Just open a new device with windows() (since your message was in a
proprietary MS character set) before each plot.
--
Brian D.
Can anyone tell me how to plot on the same graph two different functions
(of x) using two differnt y scales (axes 2 and 4)?
Thanks
Nelson
--
Use o melhor sistema de busca da Internet
Radar UOL - http://www.radaruol.com.br
Does the following produce what you want?
plot(0:1, 0:1, type=l)
axis(4, 0:1, c(0, 2))
lines(0:1, 0.5*c(2,1))
Spencer Graves
[EMAIL PROTECTED] wrote:
Can anyone tell me how to plot on the same graph two different functions
(of x) using two differnt y scales (axes 2 and 4)?
Thanks
Nelson
Hi,
Late last year there was a thread on r-help about installing Arial
fonts on linux machines:
http://finzi.psych.upenn.edu/R/Rhelp02/archive/8548.html
I too would like to use Arial fonts in some R plots created on my
redhat machine. I've installed the TrueType fonts from
Hi all,
I have a question regarding the boxplot function. The data I am working on has 1
grouping
variable (G) and it has many numerical variables (V1, V2, V3, V4, Vx, etc). What I
would like to
do is create a boxplot where the Y-axis represents the numerical values of variable
V1...Vx (all
I need a function like signif(), but returns the rounded values as
character strings, formatted with trailing zeros where appropriate.
If anyone has one, I would sure appreciate a copy.
Thanks
-Don
Details:
signif() rounds a number to a specified number of significant digits,
for example:
On Fri, 14 Mar 2003, [iso-8859-1] Rishabh Gupta wrote:
Hi all,
I have a question regarding the boxplot function. The data I am working on has 1
grouping
variable (G) and it has many numerical variables (V1, V2, V3, V4, Vx, etc). What I
would like to
do is create a boxplot where the
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Rishabh Gupta
Sent: Friday, March 14, 2003 4:10 PM
To: [EMAIL PROTECTED]
Subject: [R] boxplots with multiple numerical variables
Hi all,
I have a question regarding the boxplot function. The data
I am
Does this do what you want?
x - c(2.503,2.477,0.1204)
sapply(signif(x,3), sprintf, fmt=%#.3g)
[1] 2.50 2.48 0.120
This will give you scientific notation for very large or small numbers.
At Friday 02:17 PM 3/14/2003 -0800, Don MacQueen wrote:
I need a function like signif(), but returns
I asked why length(byyr$cnd95[byyr$cnd95==tr]) was counting NA's as well as trs.
Thanks to Brian Ripley, Uwe Ligges, Ott Toomet, Marc Schwartz, and Thomas Lumley for
their help.
Two solutions were to replace my call with
length[byyr$cnd95 %in% tr] or with sum(byyr$cnd95 == tr, na.rm=TRUE)
and
Release 1.0-rc1 of RSessionDA is now available. This version provides
significantly improved security and should now be usable on public internet
sites.
Description:
RSessionDA provides an interface to R from the open-source web application
development system Zope http://www.zope.org. This
everytime i try to read in data i get the following error:
Error in scan(file = file, what = what, sep = sep, quote = quote, dec = dec, :
line 3 did not have 5 elements
it was my understanding it wasn't necessary to specifiy all defaults. The bizarre
thing is this error comes from
Jeremy Z Butler [EMAIL PROTECTED] writes:
To creat a series of plots I've constructed the following, however
this doesn't work because the colnames being feed into n are embraced
by quotes which is stuffing up the plotting function. Is there a way
to suppress to quotes? coercion in to a
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Thomas Lumley
Sent: Friday, March 14, 2003 4:36 PM
To: Rishabh Gupta
Cc: [EMAIL PROTECTED]
Subject: Re: [R] boxplots with multiple numerical variables
On Fri, 14 Mar 2003, [iso-8859-1] Rishabh Gupta wrote:
In trying to get R CMD check to run for Hmisc and Design, I get a lot of warnings
about key words that are not defined to R. Is there now a way to add new key words to
the master list, or to have a local list that is concatenated to the master? If there
is no way to do this, I assume that the
Hi,
I just tried both of the solutions that you provided. They are perfect, exacrly what
I was
looking. Many Thanks four your help, I appreciate it.
Rishabh
--- Marc Schwartz [EMAIL PROTECTED] wrote: -Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
On Fri, 14 Mar 2003, Nirvan Sunderam wrote:
everytime i try to read in data i get the following error:
Error in scan(file = file, what = what, sep = sep, quote = quote, dec = dec, :
line 3 did not have 5 elements
it was my understanding it wasn't necessary to specifiy all
[This sort of question might be better on R-devel.]
On Fri, 14 Mar 2003, Frank E Harrell Jr wrote:
In trying to get R CMD check to run for Hmisc and Design, I get a
lot of warnings about key words that are not defined to R.
Is there now a way to add new key words to the master list, or to
55 matches
Mail list logo