[R-sig-eco] R-sig-ecology Digest, Vol 57, Issue 1

[Jade Maggs]

Thank you Tim and Martin for your advice. It is much appreciated.

Jade


On 01/12/2012 13:00, r-sig-ecology-request at r-project.org wrote:
> Send R-sig-ecology mailing list submissions to
> 	r-sig-ecology at r-project.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> 	https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
> or, via email, send a message with subject or body 'help' to
> 	r-sig-ecology-request at r-project.org
>
> You can reach the person managing the list at
> 	r-sig-ecology-owner at r-project.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of R-sig-ecology digest..."
>
>
> Today's Topics:
>
>     1. Hurdle Models: Error in	solve.default(as.matrix(fit$hessian))
>        (Yaiza Iglesias)
>     2. patterns in weather data that could relate to pathogen
>        prevalence (Anto Raja)
>     3. Re: GLM: calculate percentage deviance of factor (Martin Weiser)
>     4. Re: patterns in weather data that could relate to pathogen
>        prevalence (Bob O'Hara)
>     5. Re: GLM: calculate percentage deviance of factor (Tim Meehan)
>     6. Re: Zuur / Pinierho / Faraway (Dave Roberts)
>     7. Re: Zuur / Pinierho / Faraway (Philippi, Tom)
>     8. Re: summing F stats and permutation (Eduard Sz?cs)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Fri, 30 Nov 2012 12:40:52 +0100
> From: Yaiza Iglesias <yaiza.iglesias.ba at gmail.com>
> To: r-sig-ecology at r-project.org
> Subject: [R-sig-eco] Hurdle Models: Error in
> 	solve.default(as.matrix(fit$hessian))
> Message-ID:
> 	<CAOR0OSjDB-GDG_Uq0vmRwANZyTLihU1xxG_U_NXwc9iMtEZapA at mail.gmail.com>
> Content-Type: text/plain
>
> Hello list,
>
> I want to model the abundance of a aquatic bird using three predictors and
> I am trying to fit a Hurdle model using the R package ‘pscl’.
>
> When I try to fit a model with interaction between predictors, R give me an
> error:
>
> hurdle(formula = carbototal ~ anchomedio:distdormidero, data =
> ModHurdle300, dist = "negbin")
> Error en solve.default(as.matrix(fit_count$hessian)) :
>    system is computationally singular: reciprocal condition number = 0
>
>
> The error is the same for all models with interaction between predictors:
>
>    formula = carbototal ~ anchomedio:distdormidero
>    formula = carbototal ~ anchomedio:distpistacar
>    formula = carbototal ~ anchomedio:distdormidero:distpistacar
>
>    Being: anchomedio, river width; distdormidero, distance to the nearest
> roost and      distpistcar, distance to nearest road.
>
>
> and for the three data sets of increasing sampling size: ModHurdle100,
> where N=100; ModHurdle200, where N=200; ModHurdle300, where N=300. With the
> exception of "formula = carbototal ~ anchomedio:distpistacar" for
> ModHurdle100 data set, which the program seems able to handle without error.
>
> Is there anything I can do to solve this error? O maybe those models cannot
> be fitted?
>
> Greetings and thanks in advance for your attention.
>

-- 
_____________________________________________________________
Jade Maggs MSc
Assistant Scientist
Linefish Research
Oceanographic Research Institute
Direct Tel: +27 (31) 328 8171 Fax: +27 (31) 328 8188
Cell: +27 83 515 1079 e-mail: jmaggs at ori.org.za
1 King Shaka Avenue, Point, Durban 4001 KwaZulu-Natal South Africa
PO Box 10712, Marine Parade 4056 KwaZulu-Natal South Africa
http://www.ori.org.za/staff.html#Jade
http://www.linkedin.com/pub/jade-maggs/29/917/5a0