--- Begin Message ---
Send R-help mailing list submissions to
[EMAIL PROTECTED]
To subscribe or unsubscribe via the World Wide Web, visit
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
or, via email, send a message with subject or body 'help' to
[EMAIL PROTECTED]
You can reach the person managing the list at
[EMAIL PROTECTED]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of R-help digest..."
Today's Topics:
1. Re: Scripting with an external editor (Jim Lemon)
2. Install on Solaris 5.9 (Campbell)
3. Re: Install on Solaris 5.9 ([EMAIL PROTECTED])
4. Re: Robust standard errors (John Fox)
5. cross correlation (Martin Wegmann)
6. Help regarding C/C++ usage .. ([EMAIL PROTECTED])
7. About Scheffe test ([EMAIL PROTECTED])
8. Re: Help regarding C/C++ usage .. ([EMAIL PROTECTED])
9. Re: Help regarding C/C++ usage .. (Roger Bivand)
10. Re: Robust standard errors (Thomas Lumley)
11. AW: [R] wavelet model on time series in R? (Till Baumgaertel)
12. Is it a bug in list() behavior? (wolski)
13. Re: Is it a bug in list() behavior? (Uwe Ligges)
14. Problem with the step() function (Luis Torgo)
15. Re: Problem with the step() function ([EMAIL PROTECTED])
16. Re: Is it a bug in list() behavior? (Tony Plate)
17. cross correaltion (Martin Wegmann)
18. Re: Problem with the step() function (Thomas Lumley)
19. Box Plot Question (John Borders)
20. Re: cross correaltion ([EMAIL PROTECTED])
21. Re: Is it a bug in list() behavior? (Uwe Ligges)
22. Re: Box Plot Question (Uwe Ligges)
23. Re: Is it a bug in list() behavior? (Spencer Graves)
24. how to show a section of a matrix neatly (or row by row) that
satisfies some condition (Edmond Ng)
25. Re: Is it a bug in list() behavior? (Tony Plate)
26. RE: how to show a section of a matrix neatly (or row by row )
that satisfies some condition (Wiener, Matthew)
27. Re: how to show a section of a matrix neatly (or row by row)
that satisfies some condition (Spencer Graves)
28. Winedt and R on Windows XP (Yang, Richard)
29. Re: Winedt and R on Windows XP (Uwe Ligges)
30. compiling R on Linux w/ Intel compilers (Liaw, Andy)
31. Re: Is it a bug in list() behavior? (Uwe Ligges)
32. using tapply with a matrix? (Nurnberg-LaZerte)
33. Re: using tapply with a matrix? (Thomas Lumley)
34. Re: using tapply with a matrix? (Nurnberg-LaZerte)
35. RE: Box Plot Question (Adaikalavan Ramasamy)
36. negative binomial regression (Ross Nelson)
37. Re: negative binomial regression ([EMAIL PROTECTED])
38. isoMDS results (Christian Hennig)
----------------------------------------------------------------------
Message: 1
Date: Mon, 24 Mar 2003 22:12:14 +1100
From: Jim Lemon <[EMAIL PROTECTED]>
Subject: Re: [R] Scripting with an external editor
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="iso-8859-15"
[EMAIL PROTECTED] wrote:
> ?parse says
>
> `parse' returns the parsed but unevaluated expressions in a list.
> Each element of the list is of mode `expression'.
>
> that is, it takes all the input and returns a list once all the input
> has been parsed. There is no multi-tasking in the R evaluator, and you
> have no loop in your code, so eval can only be called once when parse
> exits.
>
> I am not clear what you actually want. If it is to repeatedly send a
> block of code to R and get that block evaluated then you need to
> indicate the end of the block somehow, split the pipe stream up into
> blocks and run a loop to parse and evaluate each block. If you want to
> simulate the command-line, the easiest way to do this is to submit to
> the command line, and to help with that you would have to at least tell
> us your platform!
My apologies.
OS - Linux (RedHat 7.2)
I had read the section in the R Language Manual about parsing:
"The read-eval-print loop forms the basic command line interface to R.
Textual input is read until a complete R expression is available."
The initial aim is to send selected text from the editor to the R command
line. If I manually invoke the editor from a terminal, I can do exactly
this, with each section of text appearing as if entered when sent from the
external editor. However, your suggestion about blocking is probably the
answer, as I noticed the blocking option but decided that the explanation
on the help page for "connections" implied that a non-blocking mode would
be the correct one:
"In non-blocking mode, operations return as soon as possible, so on input
they will return with whatever input is available..."
Any suggestions would be greatly appreciated.
Jim
------------------------------
Message: 2
Date: Mon, 24 Mar 2003 11:37:46 +0000
From: Campbell <[EMAIL PROTECTED]>
Subject: [R] Install on Solaris 5.9
To: r-help <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=US-ASCII
I'm trying to install R-1.6.2 on a Sparc machine running Solaris 5.9. I download and
unpack R, run ./Configure from the R directory. Configure fails.The last line on the
Configure script prints is 'Checking for Fortran 77 name-mangling scheme', This
output and the config.log files are available at:
http://www.phineas.pwp.blueyonder.co.uk/config.log
and
http://www.phineas.pwp.blueyonder.co.uk/configout
If you prefer I could post these 2 files directly to the mailing list.
I currently have autoconf 2.57, automake 1.7.2, gcc 3.2.2, make 3.80 and readline 4.3
installed.
Phineas Campbell
------------------------------
Message: 3
Date: Mon, 24 Mar 2003 11:59:15 +0000 (GMT)
From: [EMAIL PROTECTED]
Subject: Re: [R] Install on Solaris 5.9
To: Campbell <[EMAIL PROTECTED]>
Cc: r-help <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: TEXT/PLAIN; charset=US-ASCII
On Mon, 24 Mar 2003, Campbell wrote:
> I'm trying to install R-1.6.2 on a Sparc machine running Solaris 5.9. I download
> and unpack R, run ./Configure from the R directory. Configure fails.The last line
> on the Configure script prints is 'Checking for Fortran 77 name-mangling scheme',
> This output and the config.log files are available at:
> http://www.phineas.pwp.blueyonder.co.uk/config.log
> and
> http://www.phineas.pwp.blueyonder.co.uk/configout
>
> If you prefer I could post these 2 files directly to the mailing list.
>
> I currently have autoconf 2.57, automake 1.7.2, gcc 3.2.2, make 3.80
and readline 4.3 installed.
gcc 3.2.2 has a fatal bug for compiling R on Solaris, so the first thing
you need is a less buggy C compiler. Preferably the Sun C/F9x compilers,
but failing that gcc 3.2 will do.
You don't appear to have a working Fortran compiler installed, as
config.log says at the end (before the symbols)
configure:15266: error: cannot compile a simple Fortran program
See `config.log' for more details.
It's found something called fc, which does not work. If you have really
installed gcc 3.2.2, the build went wrong and g77 did not get installed.
--
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, UK Fax: +44 1865 272595
------------------------------
Message: 4
Date: Mon, 24 Mar 2003 07:36:47 -0500
From: John Fox <[EMAIL PROTECTED]>
Subject: Re: [R] Robust standard errors
To: Nirmala Ravishankar <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"; format=flowed
Dear Nirmala,
The object of the various HCCM estimators is to compute standard errors
that are approximately correct when the error variance in a linear model
isn't constant. I don't see the relevance to a logit model. Perhaps you can
explain further what you have in mind (or perhaps someone else is aware of
a generalization to GLMs).
A small point: hccm.default isn't meant to be called directly, but rather
through the generic function hccm. If you look at hccm.default you'll see
that it's only purpose is to report an error when hccm is called with a
non-lm object. Since glm objects inherit from lm, hccm.lm would normally be
called, but would fail for a different reason, reporting that an unweighted
lm object is required. Perhaps this is why you set weights=NULL in the call
to glm. Weights in glm, incidentally, refer to so-called prior weights --
glm still returns weights from its last iteration. The functions hccm.lm
and hccm.default are pretty simple, and you could discover all this by
looking at them.
John
At 11:36 PM 3/23/2003 -0500, Nirmala Ravishankar wrote:
>I am trying to calculate robust standard errors for a logit model. I
>installed the package "car" and tried using hccm.default, but that
>required an lm object. Is there some way to do a similar operation for a
>glm object?
>
>
>x <- hccm.default(glm(winner ~ racebl + racehi + raceas + inchi + incmed +
>edhs + edcol + edba + agec1 + agec4 + sex + margin + regla + regbay +
>regsc + libcon+ pdem + poth, data = zol, family = binomial, weights =
>NULL))
>
>Error in hccm.default(glm(winner ~ racebl + racehi + raceas + inchi + :
> requires an lm object
> >
-----------------------------------------------------
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario, Canada L8S 4M4
email: [EMAIL PROTECTED]
phone: 905-525-9140x23604
web: www.socsci.mcmaster.ca/jfox
-----------------------------------------------------
------------------------------
Message: 5
Date: Mon, 24 Mar 2003 14:09:24 +0100
From: Martin Wegmann <[EMAIL PROTECTED]>
Subject: [R] cross correlation
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="iso-8859-1"
Hello,
does anybody has experience in doing cross correlation of image data
sets from GRASS - GIS? looking for dependencies inbetween image
composites.
what would be the command for computing a cross correlation?
thanks in advance, Martin
--
Martin Wegmann
Department of Animal Ecology and Tropical Biology
Zoology III, Biocenter
Am Hubland
97074 Würzburg
Germany
0931/888-4378
[EMAIL PROTECTED]
[EMAIL PROTECTED]
-------------- next part --------------
A non-text attachment was scrubbed...
Name: wegmann.vcf
Type: text/x-vcard
Size: 329 bytes
Desc: Card for Martin Wegmann
Url :
https://www.stat.math.ethz.ch/pipermail/r-help/attachments/20030324/6ebce064/wegmann-0001.bin
------------------------------
Message: 6
Date: Mon, 24 Mar 2003 18:53:49 +0530
From: [EMAIL PROTECTED]
Subject: [R] Help regarding C/C++ usage ..
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii
Hi,
We have started to use R for our statistical based application especially
for Clustering. Clustering is one of features of the software that we
are developing. We are developing the entire product using Microsoft
Technologies VC++, VB with ODBC. I wanted to use R for performing the
Clustering and generate results. I want to call R functions in my C++
program. How can I do that ? I would appreciate if you could give me the
entire syntax for building too. It would be great help and boost for my
software development if you could send me a sample project.
Thanks in advance awaiting a early and positive response.
Regards,
Venkat Prasad M S
------------------------------
Message: 7
Date: Mon, 24 Mar 2003 08:25:39 -0500
From: [EMAIL PROTECTED]
Subject: [R] About Scheffe test
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=iso-8859-1
Dear coleagues,
I wrote a script for a multiple comparisions
like Scheffe test, according Zar, JH. Biostatiscal Analysis
pg.196..
my data is fish larvae abundance with 4 tratments
month (with 12 levels), year (with 9 levels), station
of year (with 4 levels) and area (with 3 levels)
after an anova procedure i apply this:
Scheffe<-function(mxa,mxb,ms,na,nb,k,F){
# Scheffe test for multiple comparision
# where mxa is mean of A(month, year,station or area
# mxb is some but other treatment
# ms is mean square of anova results
# na is number of A sample nb is number
# of sample B, k is degree of freedon of
# of treatments and F is critical value to
# to compare
#if S > Scrit -reject H0 else S < Scrit -accept H0
SE<-sqrt(ms*(1/na+1/nb))
S<-abs(mxb-mxa)/SE
Scrit<-sqrt((k-1)*F)
if (S > Scrit) result<-c("rej H0") else result<-c("ac H0")
return(cbind(S,Scrit,result))
}
Is there some errors??
Is possible find the F values in R??
Thanks for some help
Marcelo
__________________________________________________________________
------------------------------
Message: 8
Date: Mon, 24 Mar 2003 13:39:40 +0000 (GMT)
From: [EMAIL PROTECTED]
Subject: Re: [R] Help regarding C/C++ usage ..
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: TEXT/PLAIN; charset=US-ASCII
Are you working for a commercial organization? This reads like a request
for technical consultancy. If the `product' is for sale you need to
ensure that you can meet the requirements of the GPL which applies to R.
This is possible, in several different ways. Although R itself is free, I
am unaware of providers of free technical consultancy for it.
On Mon, 24 Mar 2003 [EMAIL PROTECTED] wrote:
> We have started to use R for our statistical based application especially
> for Clustering. Clustering is one of features of the software that we
> are developing. We are developing the entire product using Microsoft
> Technologies VC++, VB with ODBC. I wanted to use R for performing the
> Clustering and generate results. I want to call R functions in my C++
> program. How can I do that ? I would appreciate if you could give me the
> entire syntax for building too. It would be great help and boost for my
> software development if you could send me a sample project.
>
> Thanks in advance awaiting a early and positive response.
--
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, UK Fax: +44 1865 272595
------------------------------
Message: 9
Date: Mon, 24 Mar 2003 15:32:36 +0100 (CET)
From: Roger Bivand <[EMAIL PROTECTED]>
Subject: Re: [R] Help regarding C/C++ usage ..
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: TEXT/PLAIN; charset=US-ASCII
On Mon, 24 Mar 2003 [EMAIL PROTECTED] wrote:
> Hi,
>
> We have started to use R for our statistical based application especially
> for Clustering. Clustering is one of features of the software that we
> are developing. We are developing the entire product using Microsoft
> Technologies VC++, VB with ODBC. I wanted to use R for performing the
> Clustering and generate results. I want to call R functions in my C++
> program. How can I do that ? I would appreciate if you could give me the
> entire syntax for building too. It would be great help and boost for my
> software development if you could send me a sample project.
Two points: most of the information you may find useful is documenmted on
the R project website, and secondly, because R is open source, doubts
about documented functionality can be resolved by reading the entire code,
which is entirely at your disposal. I'm sure lots of people would be
interested in your generous contribution of your conclusions back to the
community. Specifically, use of the search facility on the website will
give you an initial guide to where to look - treat this as an exercise in
datamining on R.
>
> Thanks in advance awaiting a early and positive response.
>
> Regards,
> Venkat Prasad M S
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
e-mail: [EMAIL PROTECTED]
------------------------------
Message: 10
Date: Mon, 24 Mar 2003 07:19:37 -0800 (PST)
From: Thomas Lumley <[EMAIL PROTECTED]>
Subject: Re: [R] Robust standard errors
To: John Fox <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: TEXT/PLAIN; charset=US-ASCII
On Mon, 24 Mar 2003, John Fox wrote:
> Dear Nirmala,
>
> The object of the various HCCM estimators is to compute standard errors
> that are approximately correct when the error variance in a linear model
> isn't constant. I don't see the relevance to a logit model. Perhaps you can
> explain further what you have in mind (or perhaps someone else is aware of
> a generalization to GLMs).
Exactly the same reasoning as for linear models leads to `sandwich'
standard error estimators for GLMs. GLMs ordinarily require mean and
variance correctly specified to give valid standard errors; the sandwich
estimators require only that the mean is correctly specified.
For a binary logit model on independent observations this doesn't get you
much further, as the entire likelihood is going to be correct if the mean
is correct, but for other GLMs it is a useful technique.
Some options are to use the gee or geepack packages to do an analysis with
only one observation per group, or the survey package to do an analysis
without sampling weights.
-thomas
------------------------------
Message: 11
Date: Mon, 24 Mar 2003 16:31:29 +0100
From: "Till Baumgaertel" <[EMAIL PROTECTED]>
Subject: AW: [R] wavelet model on time series in R?
To: [EMAIL PROTECTED]
Cc: Yan Yu <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="ISO-8859-15"
Yan,
There's the wavethresh- and the waveslim-package.
I work with wavethresh and use lda from MASS-package to make prediction
which works quite good.
My original data is not a time-series, but a physical curve. But for your
use it shouldn't make *any* difference.
cu
till
>-- Original Nachricht --
>Date: Sun, 23 Mar 2003 23:31:57 -0800 (PST)
>From: Yan Yu <[EMAIL PROTECTED]>
>To: <[EMAIL PROTECTED]>
>Subject: [R] wavelet model on time series in R?
>
>
>Hello,
>Is there package built for R that can do model/prediction on time
>series data using wavelet?
>If not, does anyone know what popular software can do that?
>
>Thanks much,
>yan
>
>______________________________________________
>[EMAIL PROTECTED] mailing list
>https://www.stat.math.ethz.ch/mailman/listinfo/r-help
________________________________________
Mehr Power für Ihre eMail - mit den neuen Leistungspaketen bei http://www.epost.de
------------------------------
Message: 12
Date: Mon, 24 Mar 2003 16:48:28 +0100
From: "wolski" <[EMAIL PROTECTED]>
Subject: [R] Is it a bug in list() behavior?
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"
Hello!
let:
test<-1:3
list(test)
names(test)<-c("X11","X12","Y23")
>test[["Y2"]]
3
I had assumed that the names in a list are like a keys in a hash.
Therefore i thought that no value should be returned.
The behavior of:
>test["Y2"]
<NA>
NA
is as i expected.
Should it be as it is? How is the definition of [[]] and []?
/Eryk
------------------------------
Message: 13
Date: Mon, 24 Mar 2003 16:54:43 +0100
From: Uwe Ligges <[EMAIL PROTECTED]>
Subject: Re: [R] Is it a bug in list() behavior?
To: wolski <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii; format=flowed
wolski wrote:
> Hello!
>
> let:
>
> test<-1:3
> list(test)
> names(test)<-c("X11","X12","Y23")
>
>>test[["Y2"]]
>
> 3
>
> I had assumed that the names in a list are like a keys in a hash.
> Therefore i thought that no value should be returned.
>
> The behavior of:
>
>
>>test["Y2"]
>
> <NA>
> NA
>
> is as i expected.
>
>
> Should it be as it is? How is the definition of [[]] and []?
No! See "An Introduction to R", Section 6.1:
"The names of components may be abbreviated down to the minimum number
of letters needed to identify them uniquely. Thus Lst$coefficients may
be minimally specified as Lst$coe and Lst$covariance as Lst$cov."
Uwe Ligges
------------------------------
Message: 14
Date: Mon, 24 Mar 2003 17:01:11 +0000
From: Luis Torgo <[EMAIL PROTECTED]>
Subject: [R] Problem with the step() function
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"
Dear all,
I'm having some problems with using the step() function inside another
function. I think it is an environment problem but I do not know how to
overcome it. Any suggestions are appreciated.
I've prepared a simple example to illustrate my problem:
> library(MASS)
> data(Boston)
> my.fun <- function(dataset) {
+ l <- lm(medv ~ .,data=dataset)
+ final.l <- step(l)
+ }
> model <- my.fun(Boston)
Start: AIC= 1589.64
medv ~ crim + zn + indus + chas + nox + rm + age + dis + rad +
tax + ptratio + black + lstat
Df Sum of Sq RSS AIC
- age 1 0.1 11078.8 1587.6
- indus 1 2.5 11081.3 1587.8
<none> 11078.8 1589.6
- chas 1 219.0 11297.8 1597.5
- tax 1 242.3 11321.0 1598.6
- crim 1 243.2 11322.0 1598.6
- zn 1 257.5 11336.3 1599.3
- black 1 270.6 11349.4 1599.9
- rad 1 479.2 11557.9 1609.1
- nox 1 487.2 11565.9 1609.4
- ptratio 1 1194.2 12273.0 1639.4
- dis 1 1232.4 12311.2 1641.0
- rm 1 1871.3 12950.1 1666.6
- lstat 1 2410.8 13489.6 1687.3
Error in model.frame.default(formula = medv ~ crim + zn + indus + chas + :
Object "dataset" not found
Apparently the step() function is not able to find the "dataset" object which
is created inside the my.fun() function.
My system information:
> R.version
_
platform i686-pc-linux-gnu
arch i686
os linux-gnu
system i686, linux-gnu
status
major 1
minor 6.1
year 2002
month 11
day 01
language R
Thank you for any hints.
--
Luis Torgo
FEP/LIACC, University of Porto Phone : (+351) 22 607 88 30
Machine Learning Group Fax : (+351) 22 600 36 54
R. Campo Alegre, 823 email : [EMAIL PROTECTED]
4150 PORTO - PORTUGAL WWW : http://www.liacc.up.pt/~ltorgo
------------------------------
Message: 15
Date: Mon, 24 Mar 2003 17:46:37 +0000 (GMT)
From: [EMAIL PROTECTED]
Subject: Re: [R] Problem with the step() function
To: Luis Torgo <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: TEXT/PLAIN; charset=US-ASCII
Use R-devel, where this has been changed and your example works.
On Mon, 24 Mar 2003, Luis Torgo wrote:
> Dear all,
> I'm having some problems with using the step() function inside another
> function. I think it is an environment problem but I do not know how to
> overcome it. Any suggestions are appreciated.
>
> I've prepared a simple example to illustrate my problem:
>
> > library(MASS)
> > data(Boston)
> > my.fun <- function(dataset) {
> + l <- lm(medv ~ .,data=dataset)
> + final.l <- step(l)
> + }
> > model <- my.fun(Boston)
> Start: AIC= 1589.64
> medv ~ crim + zn + indus + chas + nox + rm + age + dis + rad +
> tax + ptratio + black + lstat
>
> Df Sum of Sq RSS AIC
> - age 1 0.1 11078.8 1587.6
> - indus 1 2.5 11081.3 1587.8
> <none> 11078.8 1589.6
> - chas 1 219.0 11297.8 1597.5
> - tax 1 242.3 11321.0 1598.6
> - crim 1 243.2 11322.0 1598.6
> - zn 1 257.5 11336.3 1599.3
> - black 1 270.6 11349.4 1599.9
> - rad 1 479.2 11557.9 1609.1
> - nox 1 487.2 11565.9 1609.4
> - ptratio 1 1194.2 12273.0 1639.4
> - dis 1 1232.4 12311.2 1641.0
> - rm 1 1871.3 12950.1 1666.6
> - lstat 1 2410.8 13489.6 1687.3
> Error in model.frame.default(formula = medv ~ crim + zn + indus + chas + :
> Object "dataset" not found
>
> Apparently the step() function is not able to find the "dataset" object which
> is created inside the my.fun() function.
>
> My system information:
> > R.version
> _
> platform i686-pc-linux-gnu
> arch i686
> os linux-gnu
> system i686, linux-gnu
> status
> major 1
> minor 6.1
> year 2002
> month 11
> day 01
> language R
>
> Thank you for any hints.
>
>
--
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, UK Fax: +44 1865 272595
------------------------------
Message: 16
Date: Mon, 24 Mar 2003 10:47:06 -0700
From: Tony Plate <[EMAIL PROTECTED]>
Subject: Re: [R] Is it a bug in list() behavior?
To: [EMAIL PROTECTED]
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"
As wolski/Eryk's example shows, it seems that "[[" for lists accepts abbreviations,
whereas "[" does not. Is this intended? (This is a difference from S-plus - both "["
and "[[" for lists accept abbreviations in S-plus (V6.1 for Windows at least.)
I couldn't find any mention of this difference in regards to accepting abbreviations
in either ?"[" or section 6.1 of the Introduction to R, or in the R Language Manual,
or in the R Reference Manual. [As an aside, I'd rather that the subset operators
didn't accept abbreviations at all,but ...]
The name returned by "[" for a non-existent element of a list also seems of dubious
correctness.
> list(abc=123)[["a"]]
[1] 123
> list(abc=123)["a"]
$"NA"
NULL
> list(abc=123)$a
[1] 123
> version
_
platform i386-pc-mingw32
arch i386
os mingw32
system i386, mingw32
status
major 1
minor 6.2
year 2003
month 01
day 10
language R
>
At Monday 04:54 PM 3/24/2003 +0100, you wrote:
>wolski wrote:
>>Hello!
>>let:
>>test<-1:3
>>list(test)
>>names(test)<-c("X11","X12","Y23")
>>
>>>test[["Y2"]]
>>3
>>I had assumed that the names in a list are like a keys in a hash.
>>Therefore i thought that no value should be returned.
>>The behavior of:
>>
>>>test["Y2"]
>><NA> NA
>>is as i expected.
>>
>>Should it be as it is? How is the definition of [[]] and []?
>
>No! See "An Introduction to R", Section 6.1:
>"The names of components may be abbreviated down to the minimum number of letters
>needed to identify them uniquely. Thus Lst$coefficients may be minimally specified as
>Lst$coe and Lst$covariance as Lst$cov."
>
>Uwe Ligges
>
>______________________________________________
>[EMAIL PROTECTED] mailing list
>https://www.stat.math.ethz.ch/mailman/listinfo/r-help
------------------------------
Message: 17
Date: Mon, 24 Mar 2003 18:47:56 +0100
From: Martin Wegmann <[EMAIL PROTECTED]>
Subject: [R] cross correaltion
To: "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="iso-8859-1"
Hello,
does anybody has experience in doing cross correlation of image data
sets from GRASS - GIS? looking for dependencies inbetween image
composites.
what would be the command for computing a cross correlation?
thanks in advance, Martin
--
Martin Wegmann
Department of Animal Ecology and Tropical Biology
Zoology III, Biocenter
Am Hubland
97074 Würzburg
Germany
0931/888-4378
[EMAIL PROTECTED]
[EMAIL PROTECTED]
-------------- next part --------------
A non-text attachment was scrubbed...
Name: wegmann.vcf
Type: text/x-vcard
Size: 329 bytes
Desc: Card for Martin Wegmann
Url :
https://www.stat.math.ethz.ch/pipermail/r-help/attachments/20030324/0b4d4fab/wegmann-0001.bin
------------------------------
Message: 18
Date: Mon, 24 Mar 2003 09:52:19 -0800 (PST)
From: Thomas Lumley <[EMAIL PROTECTED]>
Subject: Re: [R] Problem with the step() function
To: Luis Torgo <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: TEXT/PLAIN; charset=US-ASCII
On Mon, 24 Mar 2003, Luis Torgo wrote:
> Dear all,
> I'm having some problems with using the step() function inside another
> function. I think it is an environment problem but I do not know how to
> overcome it. Any suggestions are appreciated.
It's a bug. It's been reported and fixed for some time in the development
version.
-thomas
------------------------------
Message: 19
Date: Mon, 24 Mar 2003 12:56:21 -0500
From: John Borders <[EMAIL PROTECTED]>
Subject: [R] Box Plot Question
To: "'[EMAIL PROTECTED]'" <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain
I would like to create 15 box plots from two sets of data.
Set1 - containts PayGrade, Min_Salary, Max_Salary data for 15 pay grades
Set2 - contains PayGrade, Actual_Min, Actual_Max, and Actual_Mean for the 15
pay grades
I would like 15 box plots (one for each paygrade) whose whiskers were the
Min_Salary and Max_Salary data
and whose 'box' was Actual_Min, Actual_Mean, Actual_Max
in order to show -- for each of teh 15 grades -- the width of the
'theoretical' pay grade, and the distribution of actual salaries within each
of the 15 grades.
Any suggestions?
Thank you for any help.
J Borders (novice user)
[[alternate HTML version deleted]]
------------------------------
Message: 20
Date: Mon, 24 Mar 2003 10:15:23 -0800
From: [EMAIL PROTECTED]
Subject: Re: [R] cross correaltion
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii
Have you tried the GRASS mailing lists?
http://grass.itc.it/support.html
> Hello,
>
> does anybody has experience in doing cross correlation of image data
> sets from GRASS - GIS? looking for dependencies inbetween image
> composites.
> what would be the command for computing a cross correlation?
>
> thanks in advance, Martin
>
------------------------------
Message: 21
Date: Mon, 24 Mar 2003 19:31:37 +0100
From: Uwe Ligges <[EMAIL PROTECTED]>
Subject: Re: [R] Is it a bug in list() behavior?
To: Tony Plate <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=windows-1252; format=flowed
Tony Plate wrote:
> As wolski/Eryk's example shows, it seems that "[[" for lists accepts abbreviations,
> whereas "[" does not. Is this intended? (This is a difference from S-plus - both
> "[" and "[[" for lists accept abbreviations in S-plus (V6.1 for Windows at least.)
The general subscripting operator [] doesn't support abbreviations at
all. I don't know of any reference that states [] supports partial
matching of character strings.
> I couldn't find any mention of this difference in regards to accepting abbreviations
> in either ?"[" or section 6.1 of the Introduction to R, or in the R Language Manual,
> or in the R Reference Manual. [As an aside, I'd rather that the subset operators
> didn't accept abbreviations at all,but ...]
[[]] is the component extractor for lists, and the reference I gave
tells us that partial matching works for component indexing.
I agree that it's a good idea to mention this behaviour in the R
Language *Definition* manual.
> The name returned by "[" for a non-existent element of a list also seems of dubious
> correctness.
>
>
>>list(abc=123)[["a"]]
>
> [1] 123
>
>>list(abc=123)["a"]
>
> $"NA"
> NULL
Everything as expected from my point of view. Do you mean the "NA" is
"dubious"?
See the R Language Definition, Section 3.4.1:
"Notice however, that there are different modes of NA—the literal
constant is of mode "logical", but it is frequently automatically
coerced to other types."
Remember, it's a name!
Uwe Ligges
>
>>list(abc=123)$a
>
> [1] 123
>
>>version
>
> _
> platform i386-pc-mingw32
> arch i386
> os mingw32
> system i386, mingw32
> status
> major 1
> minor 6.2
> year 2003
> month 01
> day 10
> language R
>
>
>
> At Monday 04:54 PM 3/24/2003 +0100, you wrote:
>
>>wolski wrote:
>>
>>>Hello!
>>>let:
>>>test<-1:3
>>>list(test)
>>>names(test)<-c("X11","X12","Y23")
>>>
>>>
>>>>test[["Y2"]]
>>>
>>>3
>>>I had assumed that the names in a list are like a keys in a hash.
>>>Therefore i thought that no value should be returned.
>>>The behavior of:
>>>
>>>
>>>>test["Y2"]
>>>
>>><NA> NA
>>>is as i expected.
>>>
>>>Should it be as it is? How is the definition of [[]] and []?
>>
>>No! See "An Introduction to R", Section 6.1:
>>"The names of components may be abbreviated down to the minimum number of letters
>>needed to identify them uniquely. Thus Lst$coefficients may be minimally specified
>>as Lst$coe and Lst$covariance as Lst$cov."
>>
>>Uwe Ligges
>>
>>______________________________________________
>>[EMAIL PROTECTED] mailing list
>>https://www.stat.math.ethz.ch/mailman/listinfo/r-help
>
>
> ______________________________________________
> [EMAIL PROTECTED] mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
------------------------------
Message: 22
Date: Mon, 24 Mar 2003 19:42:37 +0100
From: Uwe Ligges <[EMAIL PROTECTED]>
Subject: Re: [R] Box Plot Question
To: John Borders <[EMAIL PROTECTED]>
Cc: "'[EMAIL PROTECTED]'" <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii; format=flowed
John Borders wrote:
> I would like to create 15 box plots from two sets of data.
>
> Set1 - containts PayGrade, Min_Salary, Max_Salary data for 15 pay grades
> Set2 - contains PayGrade, Actual_Min, Actual_Max, and Actual_Mean for the 15
> pay grades
>
> I would like 15 box plots (one for each paygrade) whose whiskers were the
> Min_Salary and Max_Salary data
> and whose 'box' was Actual_Min, Actual_Mean, Actual_Max
a) Use merge() to merge the data frames.
> in order to show -- for each of teh 15 grades -- the width of the
> 'theoretical' pay grade, and the distribution of actual salaries within each
> of the 15 grades.
>
> Any suggestions?
b) Look how boxplot.default() calls bxp() for plotting.
You might want to call bxp() directly with an appropriate structured
object (a list like that one boxplot.default() returns) as its argument.
Uwe Ligges
> Thank you for any help.
>
> J Borders (novice user)
>
> [[alternate HTML version deleted]]
>
> ______________________________________________
> [EMAIL PROTECTED] mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
------------------------------
Message: 23
Date: Mon, 24 Mar 2003 11:12:55 -0800
From: Spencer Graves <[EMAIL PROTECTED]>
Subject: Re: [R] Is it a bug in list() behavior?
To: Uwe Ligges <[EMAIL PROTECTED]>
Cc: Tony Plate <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=windows-1252; format=flowed
This is a difference between S-Plus and R.
S-Plus 6.1 for Windows Professional Ed. Rel. 1:
> tst <- c(a1 = 1, b2 = 3)
> tst["a"]
a1
1
R 1.6.2:
> tst <- c(a1=1, b2=3)
> tst["a"]
<NA>
NA
This is important for me, because some of my collaborators use S-Plus
but not R and others use R but not S-Plus. It's best for me if I can
adopt a style of use that is maximally transportable.
Best Wishes,
Spencer Graves
Uwe Ligges wrote:
> Tony Plate wrote:
>
>> As wolski/Eryk's example shows, it seems that "[[" for lists accepts
>> abbreviations, whereas "[" does not. Is this intended? (This is a
>> difference from S-plus - both "[" and "[[" for lists accept
>> abbreviations in S-plus (V6.1 for Windows at least.)
>
>
> The general subscripting operator [] doesn't support abbreviations at
> all. I don't know of any reference that states [] supports partial
> matching of character strings.
>
>
>> I couldn't find any mention of this difference in regards to accepting
>> abbreviations in either ?"[" or section 6.1 of the Introduction to R,
>> or in the R Language Manual, or in the R Reference Manual. [As an
>> aside, I'd rather that the subset operators didn't accept
>> abbreviations at all,but ...]
>
>
> [[]] is the component extractor for lists, and the reference I gave
> tells us that partial matching works for component indexing.
> I agree that it's a good idea to mention this behaviour in the R
> Language *Definition* manual.
>
>
>
>> The name returned by "[" for a non-existent element of a list also
>> seems of dubious correctness.
>>
>>
>>> list(abc=123)[["a"]]
>>
>>
>> [1] 123
>>
>>> list(abc=123)["a"]
>>
>>
>> $"NA"
>> NULL
>
>
> Everything as expected from my point of view. Do you mean the "NA" is
> "dubious"?
>
> See the R Language Definition, Section 3.4.1:
> "Notice however, that there are different modes of NA—the literal
> constant is of mode "logical", but it is frequently automatically
> coerced to other types."
> Remember, it's a name!
>
> Uwe Ligges
>
>
>>
>>> list(abc=123)$a
>>
>>
>> [1] 123
>>
>>> version
>>
>>
>> _ platform i386-pc-mingw32
>> arch i386 os mingw32 system i386,
>> mingw32 status major 1 minor
>> 6.2 year 2003 month 01
>> day 10 language R
>>
>>
>> At Monday 04:54 PM 3/24/2003 +0100, you wrote:
>>
>>> wolski wrote:
>>>
>>>> Hello!
>>>> let:
>>>> test<-1:3
>>>> list(test)
>>>> names(test)<-c("X11","X12","Y23")
>>>>
>>>>
>>>>> test[["Y2"]]
>>>>
>>>>
>>>> 3
>>>> I had assumed that the names in a list are like a keys in a hash.
>>>> Therefore i thought that no value should be returned.
>>>> The behavior of:
>>>>
>>>>
>>>>> test["Y2"]
>>>>
>>>>
>>>> <NA> NA is as i expected.
>>>>
>>>> Should it be as it is? How is the definition of [[]] and []?
>>>
>>>
>>> No! See "An Introduction to R", Section 6.1:
>>> "The names of components may be abbreviated down to the minimum
>>> number of letters needed to identify them uniquely. Thus
>>> Lst$coefficients may be minimally specified as Lst$coe and
>>> Lst$covariance as Lst$cov."
>>>
>>> Uwe Ligges
>>>
>>> ______________________________________________
>>> [EMAIL PROTECTED] mailing list
>>> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
>>
>>
>>
>> ______________________________________________
>> [EMAIL PROTECTED] mailing list
>> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
>
>
> ______________________________________________
> [EMAIL PROTECTED] mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
------------------------------
Message: 24
Date: Mon, 24 Mar 2003 19:40:06 +0000
From: "Edmond Ng" <[EMAIL PROTECTED]>
Subject: [R] how to show a section of a matrix neatly (or row by row)
that satisfies some condition
To: <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=US-ASCII
Hi all,
I want to show the rows of a matrix (100x3) of which some of its elements satisfy a
certain condition. In particular, how can I display the 2 rows of my matrix one after
the
other and not all all elements of column 1 first, then column 2 and so on. See the
following.
> finest[finest[,1] > 10^3]
[1] 4.960632e+13 3.612619e+04 7.668204e+12 1.001911e+04 8.886153e-01
[6] 6.130777e-01
> dim(finest)
[1] 100 3
Many thanks in advance.
Edmond
------------------------------
Message: 25
Date: Mon, 24 Mar 2003 12:45:18 -0700
From: Tony Plate <[EMAIL PROTECTED]>
Subject: Re: [R] Is it a bug in list() behavior?
To: Uwe Ligges <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"
At Monday 07:31 PM 3/24/2003 +0100, Uwe Ligges wrote:
>Tony Plate wrote:
>>As wolski/Eryk's example shows, it seems that "[[" for lists accepts abbreviations,
>>whereas "[" does not. Is this intended? (This is a difference from S-plus - both
>>"[" and "[[" for lists accept abbreviations in S-plus (V6.1 for Windows at least.)
>
>The general subscripting operator [] doesn't support abbreviations at all. I don't
>know of any reference that states [] supports partial matching of character strings.
My copy of the Blue Book, Section 11.4.1 (p357 of 1996 printing) seems to pretty
strongly imply that "[" supports partial matching of character strings (it gives
S-code for handling of indices, and uses pmatch for handling character indices in
extraction contexts). However, I certainly wouldn't advocate adding this to R if all
existing software works without this capability. It does seem worth documenting in
place where beginning users can find it though.
>>The name returned by "[" for a non-existent element of a list also seems of dubious
>>correctness.
>>
>>>list(abc=123)[["a"]]
>>[1] 123
>>
>>>list(abc=123)["a"]
>>$"NA"
>>NULL
>
>Everything as expected from my point of view. Do you mean the "NA" is "dubious"?
Yes, the string "NA" as a name is of dubious correctness. The behavior of "[" with
vectors is more what I would have expected:
> c(abc=123)["ab"]
<NA>
NA
>
-- Tony Plate
------------------------------
Message: 26
Date: Mon, 24 Mar 2003 14:49:41 -0500
From: "Wiener, Matthew" <[EMAIL PROTECTED]>
Subject: RE: [R] how to show a section of a matrix neatly (or row by
row ) that satisfies some condition
To: "'Edmond Ng'" <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii
You're missing a comma in the subscript. The comma tells it that your test
is for the appropriate rows.
You want: finest[finest[,1] > 10^3,]
Hope this helps,
Matt Wiener
-----Original Message-----
From: Edmond Ng [mailto:[EMAIL PROTECTED]
Sent: Monday, March 24, 2003 2:40 PM
To: [EMAIL PROTECTED]
Subject: [R] how to show a section of a matrix neatly (or row by row) that
satisfies some condition
Hi all,
I want to show the rows of a matrix (100x3) of which some of its elements
satisfy a certain condition. In particular, how can I display the 2 rows of
my matrix one after the
other and not all all elements of column 1 first, then column 2 and so on.
See the following.
> finest[finest[,1] > 10^3]
[1] 4.960632e+13 3.612619e+04 7.668204e+12 1.001911e+04 8.886153e-01
[6] 6.130777e-01
> dim(finest)
[1] 100 3
Many thanks in advance.
Edmond
______________________________________________
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
------------------------------------------------------------------------------
------------------------------
Message: 27
Date: Mon, 24 Mar 2003 12:02:05 -0800
From: Spencer Graves <[EMAIL PROTECTED]>
Subject: Re: [R] how to show a section of a matrix neatly (or row by
row) that satisfies some condition
To: Edmond Ng <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii; format=flowed
I think you need a comma after "10^3", e.g.,:
tst <- array(1:9, dim=c(3,3))
tst[tst[,1]>1,]
Spencer Graves
Edmond Ng wrote:
> Hi all,
>
> I want to show the rows of a matrix (100x3) of which some of its elements satisfy a
> certain condition. In particular, how can I display the 2 rows of my matrix one
> after the
> other and not all all elements of column 1 first, then column 2 and so on. See the
> following.
>
>
>>finest[finest[,1] > 10^3]
>
> [1] 4.960632e+13 3.612619e+04 7.668204e+12 1.001911e+04 8.886153e-01
> [6] 6.130777e-01
>
>
>>dim(finest)
>
> [1] 100 3
>
>
> Many thanks in advance.
>
> Edmond
>
> ______________________________________________
> [EMAIL PROTECTED] mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
------------------------------
Message: 28
Date: Mon, 24 Mar 2003 15:33:13 -0500
From: "Yang, Richard" <[EMAIL PROTECTED]>
Subject: [R] Winedt and R on Windows XP
To: "R-Help (E-mail)" <[EMAIL PROTECTED]>
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset="iso-8859-1"
Dear All;
I have used Winedt editor in conjunction with R on the Win2K
platform for more than a year without any problem. Recently I purchased a P4
machine with Windows XP. Following the installation of R1.62 and WinEdt 5.3
to separate directories, I copied R-WinEdt to the Plugins subdirectory under
the Winedt directory, clicked on the "install" file and edited the Rprofile
with two options as follows:
options(editor="\"c:/apps/WinEdt Team/Winedt\
-c="R-Winedt-edit\" -e = r.ini -V")
options(pager="\"c:/apps/WinEdt Team/Winedt\ -C="R-Winedt\" -e =
r.ini -V")
After I started up R and opened Winedt, there were no R script, run,
history icons shown on the WinEdt window. What did I miss ? Any suggestions?
Richard
------------------------------
Message: 29
Date: Mon, 24 Mar 2003 22:26:08 +0100
From: Uwe Ligges <[EMAIL PROTECTED]>
Subject: Re: [R] Winedt and R on Windows XP
To: "Yang, Richard" <[EMAIL PROTECTED]>
Cc: "R-Help \(E-mail\)" <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii
"Yang, Richard" wrote:
>
> Dear All;
>
> I have used Winedt editor in conjunction with R on the Win2K
> platform for more than a year without any problem. Recently I purchased a P4
> machine with Windows XP. Following the installation of R1.62
I guess you mean R-1.6.2, but which version of R-WinEdt (it's mentioned
in ReadMe.txt; please use the latest one: 1.4-1)?
> and WinEdt 5.3
> to separate directories, I copied R-WinEdt to the Plugins subdirectory under
> the Winedt directory, clicked on the "install" file and edited the Rprofile
> with two options as follows:
> options(editor="\"c:/apps/WinEdt Team/Winedt\
> -c="R-Winedt-edit\" -e = r.ini -V")
> options(pager="\"c:/apps/WinEdt Team/Winedt\ -C="R-Winedt\" -e =
> r.ini -V")
These are just the lines to turn (R-)WinEdt into the editor and pager
used by R.
If you really need these lines, you have to
a) set all required quotes,
b) escape the right quotes, and
c) specify the executable instead of the path (I guess you did the
latter assuming a default installation):
options(editor="\"c:/apps/WinEdt Team/Winedt/WinEdt\"
-c=\"R-Winedt-edit\" -e=r.ini -V")
options(pager="\"c:/apps/WinEdt Team/Winedt/WinEdt\"
-C=\"R-Winedt\" -e=r.ini -V")
Please follow the examples in ReadMe.txt.
> After I started up R and opened Winedt, there were no R script, run,
> history icons shown on the WinEdt window. What did I miss ? Any suggestions?
Have you installed R-WinEdt properly (the answer is YES, if there is a
file c:/apps/WinEdt Team/Winedt/R.ini, as mentioned in the ReadMe.txt)?
Have you created the shortcuts to start WinEdt with the R-WinEdt plugin
properly (it's mentioned in ReadMe.txt).
For further communication related to R-WinEdt, please contact its author
directly.
Uwe Ligges
[BTW: Looks like my comments at DSC2003 on the desirability of a new and
convenient installation procedure for R-WinEdt were correct.]
------------------------------
Message: 30
Date: Mon, 24 Mar 2003 16:17:36 -0500
From: "Liaw, Andy" <[EMAIL PROTECTED]>
Subject: [R] compiling R on Linux w/ Intel compilers
To: "'[EMAIL PROTECTED]'" <[EMAIL PROTECTED]>
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset=iso-8859-1
Dear R-help,
Has anyone been able to compile R on Linux using Intel's C(++) and Fortran
compilers? When I tried it (using R-devel from March 19), I get the
following error when it gets to compiling the "methods" package. Can some
one tells me where to start looking? What is "error 139"? Any help greatly
appreciated!
Cheers,
Andy
PS: This is on Mandrake 9.0, running on P3 Xeon.
make[3]: Entering directory `/home/andy/R-devel/src/library/methods'
config.status: creating src/library/methods/Makefile
make[3]: Leaving directory `/home/andy/R-devel/src/library/methods'
make[3]: Entering directory `/home/andy/R-devel/src/library/methods'
config.status: creating src/library/methods/DESCRIPTION
building package 'methods'
mkdir -p -- ../../../library/methods/R
mkdir -p -- ../../../library/methods/man
make[4]: Entering directory `/home/andy/R-devel/src/library/methods/src'
config.status: creating src/library/methods/src/Makefile
make[4]: Leaving directory `/home/andy/R-devel/src/library/methods/src'
make[4]: Entering directory `/home/andy/R-devel/src/library/methods/src'
making do_substitute_direct.d from do_substitute_direct.c
making methods_list_dispatch.d from methods_list_dispatch.c
making method_meta_data.d from method_meta_data.c
making slot.d from slot.c
making class_support.d from class_support.c
making tests.d from tests.c
make[5]: Entering directory `/home/andy/R-devel/src/library/methods/src'
make[5]: `Makedeps' is up to date.
make[5]: Leaving directory `/home/andy/R-devel/src/library/methods/src'
make[5]: Entering directory `/home/andy/R-devel/src/library/methods/src'
icc -I../../../../include -I/usr/local/include -D__NO_MATH_INLINES -mp
-fpic -g -c do_substitute_direct.c -o do_substitute_direct.o
icc -I../../../../include -I/usr/local/include -D__NO_MATH_INLINES -mp
-fpic -g -c methods_list_dispatch.c -o methods_list_dispatch.o
icc -I../../../../include -I/usr/local/include -D__NO_MATH_INLINES -mp
-fpic -g -c method_meta_data.c -o method_meta_data.o
icc -I../../../../include -I/usr/local/include -D__NO_MATH_INLINES -mp
-fpic -g -c slot.c -o slot.o
icc -I../../../../include -I/usr/local/include -D__NO_MATH_INLINES -mp
-fpic -g -c class_support.c -o class_support.o
icc -I../../../../include -I/usr/local/include -D__NO_MATH_INLINES -mp
-fpic -g -c tests.c -o tests.o
icc -shared -L/usr/local/lib -o methods.so do_substitute_direct.o
methods_list_dispatch.o method_meta_data.o slot.o class_support.o tests.o
-L../../../../bin -lR
mkdir -p -- ../../../../library/methods/libs
make[5]: Leaving directory `/home/andy/R-devel/src/library/methods/src'
make[4]: Leaving directory `/home/andy/R-devel/src/library/methods/src'
make[4]: Entering directory `/home/andy/R-devel/src/library/methods'
dumping R code in package 'methods'
make[4]: *** [../../../library/methods/R/all.rda] Error 139
make[4]: Leaving directory `/home/andy/R-devel/src/library/methods'
make[3]: *** [all] Error 2
make[3]: Leaving directory `/home/andy/R-devel/src/library/methods'
make[2]: *** [R] Error 1
make[2]: Leaving directory `/home/andy/R-devel/src/library'
make[1]: *** [R] Error 1
make[1]: Leaving directory `/home/andy/R-devel/src'
make: *** [R] Error 1
Andy I. Liaw, PhD
Biometrics Research Phone: (732) 594-0820
Merck & Co., Inc. Fax: (732) 594-1565
P.O. Box 2000, RY84-16 Rahway, NJ 07065
mailto:[EMAIL PROTECTED]
------------------------------------------------------------------------------
------------------------------
Message: 31
Date: Mon, 24 Mar 2003 23:04:35 +0100
From: Uwe Ligges <[EMAIL PROTECTED]>
Subject: Re: [R] Is it a bug in list() behavior?
To: Tony Plate <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii
Tony Plate wrote:
>
> At Monday 07:31 PM 3/24/2003 +0100, Uwe Ligges wrote:
> >Tony Plate wrote:
> >>As wolski/Eryk's example shows, it seems that "[[" for lists accepts
> >>abbreviations, whereas "[" does not. Is this intended? (This is a difference
> >>from S-plus - both "[" and "[[" for lists accept abbreviations in S-plus (V6.1 for
> >>Windows at least.)
> >
> >The general subscripting operator [] doesn't support abbreviations at all. I don't
> >know of any reference that states [] supports partial matching of character strings.
>
> My copy of the Blue Book, Section 11.4.1 (p357 of 1996 printing) seems to pretty
> strongly imply that "[" supports partial matching of character strings (it gives
> S-code for handling of indices, and uses pmatch for handling character indices in
> extraction contexts). However, I certainly wouldn't advocate adding this to R if
> all existing software works without this capability. It does seem worth documenting
> in place where beginning users can find it though.
>
> >>The name returned by "[" for a non-existent element of a list also seems of
> >>dubious correctness.
> >>
> >>>list(abc=123)[["a"]]
> >>[1] 123
> >>
> >>>list(abc=123)["a"]
> >>$"NA"
> >>NULL
> >
> >Everything as expected from my point of view. Do you mean the "NA" is "dubious"?
>
> Yes, the string "NA" as a name is of dubious correctness. The behavior of "[" with
> vectors is more what I would have expected:
> > c(abc=123)["ab"]
> <NA>
> NA
> >
>
> -- Tony Plate
Two last points:
- related to Tony Plate's mail: I don't have any S books at home (where
I am right now).
- related to Spencer Graves' mail: Transportability is not really an
issue. Or do you want to write code relying on partial matching? I
won't-or try to avoid it, at least. Example:
LL <- list(a1=1, a2=2)
LL$a # Hmmm ... partial matching can be quite dangerous!
I leave this topic open now.
Uwe
------------------------------
Message: 32
Date: Mon, 24 Mar 2003 17:00:36 -0500
From: "Nurnberg-LaZerte" <[EMAIL PROTECTED]>
Subject: [R] using tapply with a matrix?
To: "R's help mailing list" <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
I've successfully created my own function (topbot) that takes a vector as input and
used it with tapply() and an input vector (data$depth) and classification factor
(data$profile):
vols <- tapply(data$depth, data$profile, topbot)
works great.
But now I want to do something similar, except my function will take a 3 column matrix
with nrows() equal to the factor's length. tapply doesn't like this; I suspect it
converts the 3 column matrox into a vector by ignoring the DIM atribute.
I can't apply the funtion to each column separately as I need the info in all three
simultaneously for the function's results.
I could always revert to "for" loops but was looking for a more concise solution in
the spirit of "R".
Thanks in advance for any suggestions,
Bruce L.
------------------------------
Message: 33
Date: Mon, 24 Mar 2003 14:30:44 -0800 (PST)
From: Thomas Lumley <[EMAIL PROTECTED]>
Subject: Re: [R] using tapply with a matrix?
To: Nurnberg-LaZerte <[EMAIL PROTECTED]>
Cc: R's help mailing list <[EMAIL PROTECTED]>
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: TEXT/PLAIN; charset=US-ASCII
On Mon, 24 Mar 2003, Nurnberg-LaZerte wrote:
> I've successfully created my own function (topbot) that takes a vector as input and
> used it with tapply() and an input vector (data$depth) and classification factor
> (data$profile):
> vols <- tapply(data$depth, data$profile, topbot)
> works great.
>
> But now I want to do something similar, except my function will take a 3
> column matrix with nrows() equal to the factor's length. tapply doesn't
> like this; I suspect it converts the 3 column matrox into a vector by
> ignoring the DIM atribute.
>
Use by()
-thomas
------------------------------
Message: 34
Date: Mon, 24 Mar 2003 20:38:55 -0500
From: "Nurnberg-LaZerte" <[EMAIL PROTECTED]>
Subject: Re: [R] using tapply with a matrix?
To: Thomas Lumley <[EMAIL PROTECTED]>
Cc: R's help mailing list <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
** Reply to message from Thomas Lumley <[EMAIL PROTECTED]> on Mon, 24 Mar 2003
14:30:44 -0800 (PST)
>Use by()
That's what I need.
Somebody should put a link to by() in the tapply() section of the html docs.
Thanks.
------------------------------
Message: 35
Date: Tue, 25 Mar 2003 11:26:08 +0800
From: "Adaikalavan Ramasamy" <[EMAIL PROTECTED]>
Subject: RE: [R] Box Plot Question
To: "John Borders" <[EMAIL PROTECTED]>, <[EMAIL PROTECTED]>
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"
If I understand you correctly, you have the five number summary (min,
q1, median/mean, q3, max).
Have a look at the function bxp() where you can supply the summary
values as a list.
-----Original Message-----
From: John Borders [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 25, 2003 1:56 AM
To: '[EMAIL PROTECTED]'
Subject: [R] Box Plot Question
I would like to create 15 box plots from two sets of data.
Set1 - containts PayGrade, Min_Salary, Max_Salary data for 15 pay
grades Set2 - contains PayGrade, Actual_Min, Actual_Max, and Actual_Mean
for the 15 pay grades
I would like 15 box plots (one for each paygrade) whose whiskers were
the Min_Salary and Max_Salary data and whose 'box' was Actual_Min,
Actual_Mean, Actual_Max
in order to show -- for each of teh 15 grades -- the width of the
'theoretical' pay grade, and the distribution of actual salaries within
each of the 15 grades.
Any suggestions?
Thank you for any help.
J Borders (novice user)
[[alternate HTML version deleted]]
______________________________________________
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
------------------------------
Message: 36
Date: Mon, 24 Mar 2003 11:40:45 -0800
From: Ross Nelson <[EMAIL PROTECTED]>
Subject: [R] negative binomial regression
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"
I would like to know if it is possible to perform negative binomial
regression with rate data (incidence density) using the glm.nb (in
MASS) function.
I used the poisson regression glm call to assess the count of injuries
across census tracts. The glm request was adjusted to handle the data
as rates using the offset parameter since the population of census
tracts can vary by a factor of three.
eg. Call:
glm(formula = inj ~ lp + rdm, family = poisson(), data = ww,
offset = log(pop))
Deviance Residuals:
Min 1Q Median 3Q Max
-17.2779 -2.6034 -0.4519 2.0837 16.9275
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.11593 0.01482 -75.290 < 2e-16 ***
lp2 0.11569 0.01477 7.835 4.70e-15 ***
lp3 0.02374 0.01763 1.346 0.178
lp4 0.17777 0.01922 9.248 < 2e-16 ***
rdm2 -0.08810 0.01747 -5.044 4.57e-07 ***
rdm3 0.08750 0.01533 5.706 1.15e-08 ***
rdm4 0.10513 0.01518 6.925 4.35e-12 ***
---
Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1
inj and pop are interval, while lp and rdm are categorical.
A test of the dispersion indicates that the data is over dispersed, and
thus that an alternative distribution should be used.
I am not sure, however, if or how to modify the glm.nb to handle this
situation.
glm.nb(formula, ..., init.theta, link = log)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 1584 bytes
Desc: not available
Url :
https://www.stat.math.ethz.ch/pipermail/r-help/attachments/20030324/8e5a3516/attachment-0001.bin
------------------------------
Message: 37
Date: Tue, 25 Mar 2003 08:09:19 +0000 (GMT)
From: [EMAIL PROTECTED]
Subject: Re: [R] negative binomial regression
To: Ross Nelson <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: TEXT/PLAIN; charset=US-ASCII
No modification is required. The standard way in S to handle offsets is
via the offset() function, and that works in glm.nb. The offset argument
to R's glm is unnecessary.
See ?Insurance and try
glm.nb(Claims ~ District + Group + Age + offset(log(Holders)),data =
Insurance)
(which is not over-dispersed and so gives some warnings).
On Mon, 24 Mar 2003, Ross Nelson wrote:
> I would like to know if it is possible to perform negative binomial
> regression with rate data (incidence density) using the glm.nb (in
> MASS) function.
>
> I used the poisson regression glm call to assess the count of injuries
> across census tracts. The glm request was adjusted to handle the data
> as rates using the offset parameter since the population of census
> tracts can vary by a factor of three.
>
> eg. Call:
> glm(formula = inj ~ lp + rdm, family = poisson(), data = ww,
> offset = log(pop))
>
> Deviance Residuals:
> Min 1Q Median 3Q Max
> -17.2779 -2.6034 -0.4519 2.0837 16.9275
>
> Coefficients:
> Estimate Std. Error z value Pr(>|z|)
> (Intercept) -1.11593 0.01482 -75.290 < 2e-16 ***
> lp2 0.11569 0.01477 7.835 4.70e-15 ***
> lp3 0.02374 0.01763 1.346 0.178
> lp4 0.17777 0.01922 9.248 < 2e-16 ***
> rdm2 -0.08810 0.01747 -5.044 4.57e-07 ***
> rdm3 0.08750 0.01533 5.706 1.15e-08 ***
> rdm4 0.10513 0.01518 6.925 4.35e-12 ***
> ---
> Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1
>
>
> inj and pop are interval, while lp and rdm are categorical.
>
>
> A test of the dispersion indicates that the data is over dispersed, and
> thus that an alternative distribution should be used.
>
> I am not sure, however, if or how to modify the glm.nb to handle this
> situation.
>
> glm.nb(formula, ..., init.theta, link = log)
>
>
--
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, UK Fax: +44 1865 272595
------------------------------
Message: 38
Date: Tue, 25 Mar 2003 09:54:35 +0100 (CET)
From: Christian Hennig <[EMAIL PROTECTED]>
Subject: [R] isoMDS results
To: r-help-request Mailing List <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: TEXT/PLAIN; charset=US-ASCII
Hi,
this is a second try to post this to the R-help mailing list. The first one
has been rejected because of a too large attachment.
Now I ask this without attaching the data. If you want to reproduce the
results, please contact me directly to get the data.
(First mail, rejected:)
> Attached there is a 149*149 dissimilarity matrix; it is a file obtained by
> save(dm,file="dissim.Rsav").
OK, here is my question:
I worry about the reproducability of the results of isoMDS.
I try
> set.seed(5678)
> mdslinux <- isoMDS(dm,k=4)
initial value 31.071976
final value 31.071976
converged
> R.version
_
platform i686-pc-linux-gnu
arch i686
os linux-gnu
system i686, linux-gnu
status
major 1
minor 6.2
year 2003
month 01
day 10
language R
My co-worker works also with the same dissimilarity matrix and did the same
on a Windows machine (unfortunately I do not have the version data, but
it should not be too old) and got
> set.seed(5678)
> mdswin <- isoMDS(dm,k=4)
initial value 31.071976
final value 24.16980
converged
As to be expected, also the resulting MDS-configurations differ.
Initially, the cmdcsale version seems to be used, and this is
identical. BTW, I often observed that the isoMDS iteration does not change
anything (but not always)
from the cmdscale initial configuration on my machine, and I have
been somewhat sceptical more than once if this is correct.
Can all this be explained with Windows/Linux differences or what else may
happen here?
Best,
Christian
--
***********************************************************************
Christian Hennig
Seminar fuer Statistik, ETH-Zentrum (LEO), CH-8092 Zuerich (currently)
and Fachbereich Mathematik-SPST/ZMS, Universitaet Hamburg
[EMAIL PROTECTED], http://stat.ethz.ch/~hennig/
[EMAIL PROTECTED], http://www.math.uni-hamburg.de/home/hennig/
#######################################################################
ich empfehle www.boag.de
------------------------------
_______________________________________________
R-help mailing list DIGESTED
[EMAIL PROTECTED]
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
End of R-help Digest, Vol 1, Issue 117
**************************************
--- End Message ---
