[R] help with hier.part

[Petr Pikal]

Hi

although I do not know anything about hier.part package I try few 
comments

- see posting guide as it suggest to try to present a toy example 
which shows your problem

- are there some error messages or the result is not as you expect?

- what is TEMP - seems to me that you need to define it before 
hier.part() function and you did it but I am not sure if it contained 
what it should contain

- usually when something "does not work" means I made a mistake 
and I have to bore deeper to to syntax and man pages of a 
function. I swear that reading all posibble sources of information is 
worth the time if you really want to use R.

HTH
Petr


On 1 Nov 2005 at 16:12, Jeffrey Stratford wrote:

Date sent:      	Tue, 01 Nov 2005 16:12:03 -0600
From:           	"Jeffrey Stratford" <stratja at auburn.edu>
To:             	<r-help at stat.math.ethz.ch>
Subject:        	[R] help with hier.part

> R-users,
> 
> Attached is the file  (SR_use2.txt) I'd like to include and includes
> column headers.  nat_est is the response variable and is the number of
> species at a particular point.  The other variables are the
> explanatory vars (vark, var2, var1, UK, U2, U1, GK, G2, G1, PK, P2,
> P1).  
> 
> Here is Walsh's sample code for hier.part:
> 
> data(urbanwq)
> env <- urbanwq[,2,8]
> hier.part(urbanwq$lec, env, fam="gaussian", gof="Rssqu")
> 
> The code I wrote is 
> 
> library(hier.part)
> SRUSE<- read.table("F:\\GEORGIA\\species_richness\\SR_use2.txt", sep="
> ", header = TRUE, row.names = 1) TEMP<- SRUSE[2:13]
> hier.part(SRUSE$nat_est,TEMP, family="NegBin", gof="logLik", barplot=
> TRUE)
> 
> So far this doesn't work and I'd really appreciate some help.
> 
> While I have your ears, what books would one make for the clueless?  
> 
> Many thanks,
> 
> Jeff
> 
> PS. nat_est is the estimated number of species (species richness).
> Around each of the sampling points I calculated the % of different
> types of cover (pine, hardwoods, number of different covers) in three
> scales around the sampling points (1000, 200, and 100 m).  What I'm
> hoping to do with the analysis is to find the best scales and
> parameters that best predicts species richness. .  
> 
> ****************************************
> Jeffrey A. Stratford, Ph.D.
> Postdoctoral Associate
> 331 Funchess Hall
> Department of Biological Sciences
> Auburn University
> Auburn, AL 36849
> 334-329-9198
> FAX 334-844-9234
> http://www.auburn.edu/~stratja
> ****************************************
> 

Petr Pikal
petr.pikal at precheza.cz