Your tests are of problems where you really should be using an optimized
BLAS. But because those pointers are twice the size, the L1 cache will
hold half as many and so I am not surprised at a factor of three on a
naive implementation.
For linear algebra on large matrices the key to good
On Fri, 10 Jun 2005, Liaw, Andy wrote:
I'm not familiar with Solaris, so take this with appropriate dose of NaCl...
For the 64-bit build, why not have the -O2 for gcc, since you have it for
g77 and g++? If you just run vanilla configure for the 32-bit build, I
believe it uses -O2 for all
Thanks for your response.
OK, multinom() is a more logical in this context.
But similar problem occurs:
Let these data to be analyzed using classical glm with binomial error:
m f factor m theo f theo
-Ln L model-Ln L full interecept
f
10 12 1.2
On Fri, 10 Jun 2005, Mike R wrote:
i've search the email archives, searched the documention
of various map packages and done an R-site search, but
have been unable to find direct resources for creating maps
of the US that are colored or annotated or ... by zipcode
data.
For example,
Hi
On 10 Jun 2005 at 20:05, Sander Oom wrote:
Dear all,
Dimitris and Andy, thanks for your great help. I have progressed to
the following code which runs very fast and effective:
mat - matrix(sample(-15:50, 15 * 10, TRUE), 15, 10)
mat[mat45] - NA
mat-NA
By this you redefine mat as
Hi Gregor,
The compiled list of genetics packages is a great idea, I'll certainly find it
useful as it took me quite some time to track them all down through CRAN.
I've a couple of things that could be added...
qvalue - A procedure for false discovery rate control. General, but developed
for
Hi
On 10 Jun 2005 at 13:58, Stefan Wagner wrote:
I am really sorry, that the code appears in a bad way but it is really
not my fault (at least I think so). I think my mail-programm is
responsible for it because I sent it in a normal way without havin g
trouble with my space bar. Here are the
Dear Valued Customer
Thank you for contacting Domainz. Your email enquiry has been received and
will be attended by a member of our Customer Service Team.
Please note that this is an system generated email response to confirm the
receipt of your request, if you need to contact us urgently
Dear Marc,
I get the same results -- same coefficients, standard errors, and fitted
probabilities -- from multinom() and glm(). It's true that the deviances
differ, but they, I believe, are defined only up to an additive constant:
dt
output factor n
1 m1.2 10
2 f1.2 12
3
Dear list,
Following off-list discussion with Graham Smith (who initiated the
initial request for a task view on this topic) and Philippe Grosjean I
have produced a Task View for ecological and environmental data
analysis. Graham and Philippe commented on a earlier draft and it is now
in a
hi ppl,
this somethin very urgent. plz anybody tell me for my problem:-
how to make a module/lib that will allow to call easily R code/functions
from C++ (C++ Builder 6). Is it possible without using any intermediate
things.
plz help
srry no time for RTFM. Any idea or hint is
On Sat, 11 Jun 2005, John Fox wrote:
Dear Marc,
I get the same results -- same coefficients, standard errors, and fitted
probabilities -- from multinom() and glm(). It's true that the deviances
differ, but they, I believe, are defined only up to an additive constant:
Yes. There are many
This is discussed in detail in the R-admin manual for R 2.1.0 (the current
version). There is no point in repeating the whole discussion (several
pages) here: that is the definitive account and you do need the whole
picture.
On Sat, 11 Jun 2005, SUMIT MALHOTRA wrote:
hi ppl,
this somethin
Dear all,
I'm new using R and in (geo)statistics. I have a problem with solving
my homework questions. We are working with variograms and trying to
write down basic equations for different models (spherical,
exponential, Gaussian). I tried to use the 'gstat' and 'geoR' packages
to solve the
Dear All,
Is there any existing package or direct function in R
/ S-plus that calculates moments (raw or central); the
usual measures of skewness and kurtosis, and/or
pearsonian k criterion?
Thank you for your time.
--
Mohammad Ehsanul Karim
Web:
dwfu wrote:
Dear all,
I'm new using R and in (geo)statistics. I have a problem with solving
my homework questions. We are working with variograms and trying to
write down basic equations for different models (spherical,
exponential, Gaussian). I tried to use the 'gstat' and 'geoR' packages
Have you tried RSiteSearch(skewness) and RSiteSearch(kurtosis)?
The fBasics package has functions for that, as do other packages.
spencer graves
Mohammad Ehsanul Karim wrote:
Dear All,
Is there any existing package or direct function in R
/ S-plus that calculates
Mohammad Ehsanul Karim wrote:
Dear All,
Is there any existing package or direct function in R
/ S-plus that calculates moments (raw or central); the
usual measures of skewness and kurtosis, and/or
pearsonian k criterion?
See package e1071.
Uwe Ligges
Thank you for your time.
Andy, Prof. Ripley - thanks for your replies.
CFLAGS was not specified, so should default to -g -O2. It is definitely
worth checking, although almost all the time in these tests will be spent
in Fortran code.
Yes - I verified that's the default.
neither build uses a BLAS.
Well, they do,
This is true, they are equal. I hadn't noticed that. Thank you.
Now, if lda fails on this given input (equal means), shouldn't we
catch it and give a slightly better error message? I've spent a good
while going through the debugging process with lda.default. From that
perspective it appears that
Dear R-helpers,
I am learning about combination in R.
I want to combination all of
possible variable but it limited.
I am sorry I could not explain exactly.
For usefull I give an example
interface - c(usb,fireware,infra,bluetooth)
screen- c(lcd,cube)
computer - c(pc,server,laptop)
--- Joshua Gilbert [EMAIL PROTECTED] wrote:
If this isn't the place to do so, tell me. But, I'd like to petition
to alter the code of lda.default.
It seems to me that if you want to alter the code of lda.default, you
have everything you need to do so. The code is there, it is GPL'd and
you
On Sat, 2005-06-11 at 20:44 +0200, Muhammad Subianto wrote:
Dear R-helpers,
I am learning about combination in R.
I want to combination all of
possible variable but it limited.
I am sorry I could not explain exactly.
For usefull I give an example
interface -
On Sat, 11 Jun 2005, John Fox wrote:
Dear Marc,
I get the same results -- same coefficients, standard errors, and fitted
probabilities -- from multinom() and glm(). It's true that the deviances
differ, but they, I believe, are defined only up to an additive constant:
Yes. There are many
Dear Marc,
I get the same results -- same coefficients, standard errors, and fitted
probabilities -- from multinom() and glm(). It's true that the deviances
differ, but they, I believe, are defined only up to an additive constant:
predict(dt.b, type=response)
1 2 3
On 6/11/05, Marc Schwartz [EMAIL PROTECTED] wrote:
On Sat, 2005-06-11 at 20:44 +0200, Muhammad Subianto wrote:
Dear R-helpers,
I am learning about combination in R.
I want to combination all of
possible variable but it limited.
I am sorry I could not explain exactly.
For usefull I
Scott Gilpin [EMAIL PROTECTED] writes:
Andy, Prof. Ripley - thanks for your replies.
CFLAGS was not specified, so should default to -g -O2. It is definitely
worth checking, although almost all the time in these tests will be spent
in Fortran code.
Yes - I verified that's the default.
Dear Marc,
-Original Message-
From: Marc Girondot [mailto:[EMAIL PROTECTED]
Sent: Saturday, June 11, 2005 2:16 PM
To: John Fox
Cc: [EMAIL PROTECTED]
Subject: RE: [R] Problem with multinom ?
Dear Marc,
I get the same results -- same coefficients, standard errors, and
fitted
The code will be changed to give a more informative error message in a
future release. However, do remember that this is volunteer code, and it
is not reasonable to expect volunteers to anticipate that a user will
apply it in extremely unlikely circumstances.
If you have a suggestion about
I use predict for predictions from glm. I am wondering if there is a
predict function for predictions from the results of GLMM model?
Thanks ahead!
Weihong Li
Undergraduate Student in Statistics
University of Alberta
__
R-help@stat.math.ethz.ch
On Sat, 11 Jun 2005, Peter Dalgaard wrote:
Scott Gilpin [EMAIL PROTECTED] writes:
Andy, Prof. Ripley - thanks for your replies.
CFLAGS was not specified, so should default to -g -O2. It is definitely
worth checking, although almost all the time in these tests will be spent
in Fortran
On 6/11/05, Prof Brian Ripley [EMAIL PROTECTED] wrote:
For the record, I timed 1000x1000 not 3000x3000 (and said so). I was not
proposing to spend several hours running timings at ca 2200s each, not
least as I used a public machine with a ban on running long jobs (we have
other much faster
I fully understand that this is a volunteer project, I'm a Debian user
(not a developer... yet).
I have read the posting guide, but I forgot the protocol. First
offence, won't happen again.
Thanks.
On 6/11/05, Prof Brian Ripley [EMAIL PROTECTED] wrote:
The code will be changed to give a more
From: Peter Dalgaard
[snip]
While your setup is in place, you might want to play around with the
higher optimization levels. GCC on AMD64 sees a quite substantial
speedup from -O2 to -O3.
On our SLES8 amd64 boxes, I had trouble with g77 -O3 (build failed). Have
not tried with newer GCC.
Spencer Graves wrote:
How might you fit a generalized linear model (glm) with variance
= mu+theta*mu^2 (where mu = mean of the exponential family random
variable and theta is a parameter to be estimated)?
This appears in Table 2.7 of Fahrmeir and Tutz (2001)
Multivariate
35 matches
Mail list logo
