On Thu, 3 Jul 2008, Ben Bolker wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
~ turns out I don't need to look at the C code.
~ if one uses the mu/size parameterization of the
negative binomial, R computes size/(size+mu) to
switch parameterizations. If size>>mu this
gets rounded to 1
> "BDR" == Prof Brian Ripley <[EMAIL PROTECTED]>
> on Fri, 4 Jul 2008 09:02:21 +0100 (BST) writes:
BDR> On Thu, 3 Jul 2008, Ben Bolker wrote:
>> -BEGIN PGP SIGNED MESSAGE- Hash: SHA1
>>
>>
>> ~ turns out I don't need to look at the C code.
>>
>> ~
On Fri, 4 Jul 2008, Martin Maechler wrote:
"BDR" == Prof Brian Ripley <[EMAIL PROTECTED]>
on Fri, 4 Jul 2008 09:02:21 +0100 (BST) writes:
BDR> On Thu, 3 Jul 2008, Ben Bolker wrote:
>> -BEGIN PGP SIGNED MESSAGE- Hash: SHA1
>>
>>
>> ~ turns out I don't need to look at
Jay,
thanks for your suggestion and for the effort you already put into this.
I have had it in my mind for some time now that a Task View related to
R and education might be a good thing.
When I read your subject line, I thought: Yes, good idea. But reading
through your suggestions I realize
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
~ thanks!
~ Poisson+warning seems very sensible.
~ Looks like things start to get wonky around size=5e14
(this is for mu=1)
but thinking a little harder about the numerical analysis
will probably give a more precise criterion.
curve(dnbinom(1,mu=1
Dear Achim,
On Fri, Jul 4, 2008 at 6:59 AM, Achim Zeileis
<[EMAIL PROTECTED]> wrote:
>
> When I read your subject line, I thought: Yes, good idea. But reading
> through your suggestions I realized that it might a bit difficult to define
> what should go in there and what not.
Yes, agreed.
>
>
> "BB" == Ben Bolker <[EMAIL PROTECTED]>
> on Fri, 04 Jul 2008 08:43:21 -0400 writes:
BB> -BEGIN PGP SIGNED MESSAGE-
BB> Hash: SHA1
BB> ~ thanks!
BB> ~ Poisson+warning seems very sensible.
better than now, in the extreme cases,
but far from optimal (in thos
Having just started using R, after almost 30 years writing my own code (in
fortran or C++) to do stats as needed, I don't really know what these "Task
Views" are, but from what I can infer from what the two of you have written,
it seems to me that the beginning of the process to produce an educatio
Are there any packages that help with statistical analysis in situations where
fractal geometry is relevant? Perhaps something that supports computation of
fractal related statistics, such as the Hurst exponent or fractal dimension?
Or perhaps which support obvious tasks such as taking samples
A new version of the 'snow' package for parallel computing in R is
available at
http://www.stat.uiowa.edu/~luke/R/cluster/snow_0.3-3.tar.gz
This substantially revises the way in which worker processes are
started to allow snow to be used on Windows and Mac/Windows/Linux
combinations. I have
In the code below test2() gives an error message:
Error in .subset2(x, i, exact = exact) :
attempt to select less than one element
even though test(), which is nearly the same, gives
the expected result. BOD is a data set that comes
with R. Is this a bug?
idx <- 2
# returns expected
RSiteSearch('fractal', 'fun') returned 68 hits for me just now.
Might any of those be helpful?
Spencer
Ted wrote:
Are there any packages that help with statistical analysis in situations where
fractal geometry is relevant? Perhaps something that supports computation of
fractal re
Dear Gabor:
Does the following answer your question:
> tst <- list(a=1, b=2)
> tst[numeric(0)]
list()
> tst[[numeric(0)]]
Error in tst[[numeric(0)]] : attempt to select less than one element
Spencer
Gabor Grothendieck wrote:
In the code below test2() gives an error message:
Er
Not really. The point of this is that promises seem to be
screwing up in conjunction with [[ but in light of the fact
that .subset2 works it suggests that there might be a better
implementation of [[.
Also some insight into what is really driving this problem
so one can more easily recognize it i
14 matches
Mail list logo
