Thank you to all those who helped me on the Zero_inflated Poisson
Distribution.
In this e-mail I summarize the answers that I got with the query about
zero-inflated Poisson distribution. What I obtained is that SAS uses a
procedure called sas proc traj there’s a very good page that gives examples
http://www.stat.cmu.edu/~bjones
Also stata has implemented this function and a good web page giving examples
of this use is:
http://www.gseis.ucla.edu/courses/ed231c/notes1/zip.html
Finally a good paper on ZIP is Multivariate Zero-Inflated Poisson Models and
Their Applications. Li et al. Technometrics.
Kind regards, Pep.
-----------------------------
This problem is solved by the EM algorithm (Dempster, Laird and Rubin, 1977,
JRSS B),
Aitkin and Wilson (1982). I do not think there is any software for doing
exactly this,
but the algorithm is relatively easy to program -- that is the point of the
EM
algorithm.
Nick Longford, DMU Leicester, UK
--------------------------------
Pep,
Para analizar datos con Zero Inflated Poisson puedes utilizar Stata:
instruccion ZIP.
Además en la versión en castellano del libro de GREENE sopbre
econometría viene un capitulo dedicado a regresión de Poisson donde
explica muy claramente todo del tema de los Zero Inflated Models.
Suerte!.
AURELIO.
--------------------------------------
Pep,
if I recall correctly, there has been some discussion of fitting zero-
inflated poisson models using SAS in the SAS-L mailing list in the
last year or so. You might try to search the archives at
http://www.listserv.uga.edu/archives/sas-l.html
Hope this helps,
Sabine
--------------------------------------
Others might suggest this too, but if you want to be Bayesian, then you
could use WinBugs (<http://www.mrc-bsu.cam.ac.uk/bugs/welcome.shtml>).
I wrote some code for this very problem this week, which I could send to
you should you decide to go down this path.
Good luck!
Bob
--
Bob O'Hara
Rolf Nevanlinna Institute
P.O. Box 4 (Yliopistonkatu 5)
FIN-00014 University of Helsinki
Finland
tel: +358 9 191 23743 mobile: +358 50 599 0540
fax: +358 9 191 22779 email: [log in to unmask]
It is being said of a certain poet, that though he tortures the English
language, he has still never yet succeeded in forcing it to reveal his
meaning
- Beachcomber
--------------------------------------------------------------------
Stata does it.
Edmond
-----------------------------------------------------------------------
Hi,
Such type of models can be fitted using finite mixtures. There is free
software available based on work by Dankmar Bohning in Berlin (Germany). De
package is called C.A.MAN which can be obtained from
http://ftp.ukbf.fu-berlin.de/sozmed/caman.html
best regards,
Geert Verbeke
Biostatistical Centre
K.U.Leuven
Kapucijnenvoer 35
B-3000 Leuven
Belgium
Tel: +32 16 33 68 91
Tel (sec): +32 16 33 68 92
Fax: +32 16 33 69 00
Web: http://www.med.kuleuven.ac.be/biostat/
_________________________________________________________________
Add photos to your messages with MSN 8. Get 2 months FREE*.
http://join.msn.com/?page=features/featuredemail
|