R help - Dec 2009 - how can generate from trunceted gamma distribution in R ?

Hi, all
How can generate a sample from truncated inverse  gamma distribution in R?

thanks

One approach is to sample from a non-truncated inverse gamma, then if the
observation is in the part you want truncated, throw it away and generate a new
value.

-- 
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.snow at imail.org
801.408.8111

> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of khazaei at ceremade.dauphine.fr
> Sent: Friday, December 11, 2009 5:12 AM
> To: r-help at r-project.org
> Subject: [R] how can generate from trunceted gamma distribution in R ?
> 
> 
> Hi, all
> How can generate a sample from truncated inverse  gamma distribution in
> R?
> 
> thanks
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.

On 11/12/2009 7:12 AM, khazaei at ceremade.dauphine.fr
wrote:
> Hi, all
> How can generate a sample from truncated inverse  gamma distribution in R?
Using the inverse CDF method or rejection sampling are possible, 
depending on what your truncation is like.  If your truncation forces 
the observations far out into the tails, you need to be careful about 
rounding and underflow when using the the inverse CDF method.

Duncan Murdoch
> 
> thanks
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

Duncan Murdoch wrote:
> On 11/12/2009 7:12 AM, khazaei at ceremade.dauphine.fr wrote:
>> Hi, all
>> How can generate a sample from truncated inverse  gamma distribution in
R?
> 
> Using the inverse CDF method or rejection sampling are possible, 
> depending on what your truncation is like.  If your truncation forces 
> the observations far out into the tails, you need to be careful about 
> rounding and underflow when using the the inverse CDF method.
> 
> Duncan Murdoch
> 
I think perusal of this paper might be a good idea:

Sampling Truncated Normal, Beta, and Gamma Densities
Paul Damien and Stephen G. Walker
Journal of Computational and Graphical Statistics, Vol. 10, No. 2 (Jun., 
2001), pp. 206-215

Remembering that the inverse gamma is the inverse of a gamma, you may be 
able to get a truncated inverse gamma from a truncated (at the other 
end) gamma. Alternatively, the methodology outlined in the paper most 
likely can be modified for the inverse gamma.

David Scott


_________________________________________________________________
David Scott	Department of Statistics
		The University of Auckland, PB 92019
		Auckland 1142,    NEW ZEALAND
Phone: +64 9 923 5055, or +64 9 373 7599 ext 85055
Email:	d.scott at auckland.ac.nz,  Fax: +64 9 373 7018

Director of Consulting, Department of Statistics