We have no position about Fortran 90. We provide the means to specify
an F90 compiler and note that its use is not portable as not all
platforms have one. (A very few are still using gcc3 with g77.)
This simply is not an R issue, and seems specific to your unstated
platform (*PLEASE* do foll
Thanks Peter. I have just a couple of minor comments, and another
possible feature request, although it's one that I don't think will be
implemented.
peter dalgaard writes:
> On Mar 14, 2011, at 22:25 , Brett Presnell wrote:
>
>>
>> Is there any reason that rstandard.glm doesn't have a "pears
On Mar 14, 2011, at 22:25 , Brett Presnell wrote:
>
> Is there any reason that rstandard.glm doesn't have a "pearson" option?
> And if not, can it be added?
Probably... I have been wondering about that too. I'm even puzzled why it isn't
the default. Deviance residuals don't have quite the prop
My apologies. I guess it would help if I tried the code more than once
before posting. That should have been:
rstandard.glm <-
function(model,
infl=influence(model, do.coef=FALSE),
type=c("deviance", "pearson"), ...)
{
type <- match.arg(type)
res <- switch(type, pear
Hi there,
I am preparing a package based on Fortran 90 code. I've put the main code
is a MODULE, and I'm having several error messages with the linker:
>R CMD SHLIB --output=mylib.so nrtype.f90 topbalmod.f90 runmodels.f90
The several error messages are related with the addition of an underscore
Is there any reason that rstandard.glm doesn't have a "pearson" option?
And if not, can it be added?
Background: I'm currently teaching an undergrad/grad-service course from
Agresti's "Introduction to Categorical Data Analysis (2nd edn)" and
deviance residuals are not used in the text. For now I
Indeed! I added the drop.unused.levels argument and it now works. Thank you
Bill!
Here is a patch which one should use at one's own risk as it redefines
rlm.formula as global object, i.e. rlm.formula is no longer in the MASS
namespace and this is certainly not a good idea:
rlm.formula <- funct
Hi there,
Thanks, but I'm using R 2.12.2(2011-02-25) and ?textConnection says
everything you quoted except the last sentence ("'object' should be the name
of a character vector: however, short expressions will be accepted provided
they deparse to less than 60 bytes."). Can it be that the help
> -Original Message-
> From: r-devel-boun...@r-project.org
> [mailto:r-devel-boun...@r-project.org] On Behalf Of Vadim Ogranovich
> Sent: Monday, March 14, 2011 10:37 AM
> To: 'r-devel@r-project.org'
> Subject: [Rd] discrepancy between lm and MASS:rlm
>
> Dear R-devel,
>
> There seems t
On Mon, 2011-03-14 at 12:52 -0400, John Fox wrote:
> Dear Terry,
>
> Possibly I'm missing something, but since the generic drop1() doesn't have a
> test argument, why is there a problem?
>
> > args(drop1)
> function (object, scope, ...)
>
> If you use match.arg() against test, then the error
Dear R-devel,
There seems to be a discrepancy in the order in which lm and rlm evaluate their
arguments. This causes rlm to sometimes produce an error where lm is just fine.
Here is a little script that illustrate the issue:
> library(MASS)
> ## create data
> n <- 100
> dat <- data.frame(x=rep(
Dear Terry,
Possibly I'm missing something, but since the generic drop1() doesn't have a
test argument, why is there a problem?
> args(drop1)
function (object, scope, ...)
If you use match.arg() against test, then the error message should be
informative if one of the prescribed values isn't s
A recent question in r-help made me realize that I should add a drop1 method
for coxph and survreg. The default does not handle strata() or cluster()
properly.
However, for coxph the right options for the "test" argument would be
likelihood-ratio, score, and Wald; not chisq and F. All of th
On Mon, 14 Mar 2011, Sigbert Klinke wrote:
Hi,
I get the same error as mentioned in
http://r.789695.n4.nabble.com/error-in-memCompress-in-make-check-td3318922.html
I try to compile R 2.12.2 under Tinycore 3.5.1 running in VirtualBox 4.0.4.
Since first xz was not installed I must have used th
Hi,
I get the same error as mentioned in
http://r.789695.n4.nabble.com/error-in-memCompress-in-make-check-td3318922.html
I try to compile R 2.12.2 under Tinycore 3.5.1 running in VirtualBox 4.0.4.
Since first xz was not installed I must have used the internal version
which comes with the R sour
Kenn,
you might have had a point a while ago, but you may want to check most recent R
and re-evaluate:
?textConnection
[...]
object: character. A description of the connection. For an input
this is an R character vector object, and for an output
connection the name for th
Hello,
`textConnection` prepares arguments for an internal function, and one of
these arguments is "description" that must be a character vector of length 1
(or so it seems).
Now the one and only argument you usually give to `textConnection` is
called "object"; from the code you can see how this
17 matches
Mail list logo
