R help - Mar 2004 - Could someone email me with the code for glm.nb ?

Hi -- subject says all.  I just want the code for that function,
which I guess was in Venables and Ripley as early as 1994.
Well, and for any of the sub-functions that glm.nb calls.  I
can't install the entire MASS library.

If the code for just glm.nb (again, don't want to touch the MASS 
library, last time I tried to install it was a complete nightmare and
fiasco) is somewhere on a webpage or ftp site where I could copy it, 
pls.  just let me know where?

Many thanks -julii

On Sun, 28 Mar 2004 J.Brainard at uea.ac.uk wrote:
> Hi -- subject says all.  I just want the code for that function,
> which I guess was in Venables and Ripley as early as 1994.
Yes, for S: that version does not work in R.
> Well, and for any of the sub-functions that glm.nb calls.  I
> can't install the entire MASS library.
If you are using R, you do not need to.  It comes with MASS installed.
If you are using S-PLUS, you have asked in the wrong place for the wrong 
version ....

Actually you need a lot more than the code you asked for, for example the 
print and anova methods.
> If the code for just glm.nb (again, don't want to touch the MASS 
> library, last time I tried to install it was a complete nightmare and
> fiasco) 
Perhaps you need to invest in some basic computer help or training?  
Especially if you somehow managed to install R without it.
> is somewhere on a webpage or ftp site where I could copy it, 
> pls.  just let me know where?
The source code for the MASS *package* for R is on CRAN in the VR bundle, 
and is part of a standard R distribution.


Readers: please do not email him the code, anyone: the package does have a
licence and should not be split up.

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
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

> Hi -- subject says all.  I just want the code for that function,
> which I guess was in Venables and Ripley as early as 1994.
> Well, and for any of the sub-functions that glm.nb calls.  I
> can't install the entire MASS library.
	Why on earth not?  What is the matter with your computer/system?
> If the code for just glm.nb (again, don't want to touch the MASS 
> library, last time I tried to install it was a complete nightmare and
> fiasco) is somewhere on a webpage or ftp site where I could copy it, 
> pls.  just let me know where?
	It's called CRAN!  If you really (and let me repeat, this is
	***weird***) don't want to install MASS, then just get the
	package source, untar it, go to the R directory, and copy
	whatever code you want to wherever you want it.

	Life would be much easier for you if you'd get over your
	hang-ups and just install MASS, but.

	If you are using Linux it's completely transparent and
	automatic.  Even if you are using Windoze (God help you!)
	it's ***still*** (mirabile dictu) completely transparent and
	automatic.

	You can do it from within R --- just execute

	> install.packages("MASS",lib=whatever)

	and R does all the rest.

					cheers,

						Rolf Turner
						rolf at math.unb.ca

It is a pity that both times I have tried to use R  and the R-mailing
list that I received unpleasant replies, esp. coming from Ripley
himself this time.  Last time I simply asked "Why would I want to
use R over Splus??"  Which provoked some apologies that I wasn't to 
know that this would touch a nerve, in addition to a volume of ranting
and raving.  At least Ripley's reply wasn't vitriolic.  
 
I did not install R.  I wouldn't even try.  It's available centrally 
running on a Unix machine.  Trying to access glm.nb() at the R prompt
leads to these replies from R:
>glm.nb
Error: "glm.nb" Object not found
>

Which is why I though I could just ask for the simple text
function code, type it in, correct typos, make it work, fool-proof, 
low hassle....  Hahahahahahahhaa.

I don't have root priveleges and unashamedly
admit to not having the ability to install either MASS or the
negbin shell archive library (I spent many frustrating failed hours last
time I tried to do that).  I didn't realise R was only for software 
whizzes.  Now I know!


Investment in people is not my employer's strong suit.  Seems to come
with university environments.  Some of us aren't that important -- except
as convenient targets for being belittled, of course.

Maybe the employer will fork out for me to learn Stata and Limdep 
instead, eh?  At least the user-support-groups & might be friendlier.  
I had foolishly thought that the R community, having an ethos of self-
sufficiency, might be friendlier and more responsive than going 
via the Splus list.  Boy am I stupid or what?

-Julii

Dear Dr. Brainard,

In addition to what Roger has pointed out, I would like to point out one
fact that you seem not to be aware of.  Open source software, by and large,
have licences, just as the proprietary ones.  Openly asking others to send
you the code, IMHO, is plain disregard of the licence for which the software
is covered under.  In this case all you received seems to be bruised ego,
but in other situation you might be visited by law enforcement, or a letter
from some attorney, neither of which I doubt you will find `friendlier'.  I
do not know how user-contributed code in Stata are licenced, but that's for
you to investigate.  Please do not confuse open source with public domain.

Also, there is nothing to prevent you from installing add-on packages
without root priviledge.  You can install packages anywhere that you have
read/write access.  The help pages for install.packages() and library() have
all the details, and I am quite sure you do not require assistance in
reading them.

With regard to your previous question: Given the tone that you asked the
question in, the most polite response I can squeeze out of myself is `Why
would I care to tell you?'  Have you tried asking your thesis advisor why
you should read his or her papers?

Best Regards,
Andy Liaw, PhD
Biometrics Research      PO Box 2000, RY33-300     
Merck Research Labs           Rahway, NJ 07065
mailto:andy_liaw at merck.com        732-594-0820


> -----Original Message-----
> From: r-help-bounces at stat.math.ethz.ch 
> [mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of 
> J.Brainard at uea.ac.uk
> Sent: Monday, March 29, 2004 1:47 AM
> To: r-help at stat.math.ethz.ch
> Subject: Re: [R] Could someone email me with the code for glm.nb ?
> 
> 
> 
> It is a pity that both times I have tried to use R  and the R-mailing
> list that I received unpleasant replies, esp. coming from Ripley
> himself this time.  Last time I simply asked "Why would I want to
> use R over Splus??"  Which provoked some apologies that I wasn't
to
> know that this would touch a nerve, in addition to a volume of ranting
> and raving.  At least Ripley's reply wasn't vitriolic.  
>  
> I did not install R.  I wouldn't even try.  It's available
centrally
> running on a Unix machine.  Trying to access glm.nb() at the R prompt
> leads to these replies from R:
> 
> >glm.nb
> Error: "glm.nb" Object not found
> >
> 
> 
> Which is why I though I could just ask for the simple text
> function code, type it in, correct typos, make it work, fool-proof, 
> low hassle....  Hahahahahahahhaa.
> 
> I don't have root priveleges and unashamedly
> admit to not having the ability to install either MASS or the
> negbin shell archive library (I spent many frustrating failed 
> hours last
> time I tried to do that).  I didn't realise R was only for software 
> whizzes.  Now I know!
> 
> 
> Investment in people is not my employer's strong suit.  Seems to come
> with university environments.  Some of us aren't that 
> important -- except
> as convenient targets for being belittled, of course.
> 
> Maybe the employer will fork out for me to learn Stata and Limdep 
> instead, eh?  At least the user-support-groups & might be 
> friendlier.  
> I had foolishly thought that the R community, having an ethos of self-
> sufficiency, might be friendlier and more responsive than going 
> via the Splus list.  Boy am I stupid or what?
> 
> -Julii
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! 
> http://www.R-project.org/posting-guide.html
> 
>