Re: [R] Overdispersion in count data

2008-04-03 Thread Wade Wall
Alright, I feel stupid now. That was the problem. For glm you can use both successes and failures, while with the negative binomial it is simply a count. That is why I was getting the subscript too long message. I understand generalized linear models, but I haven't worked with negative binomial

Re: [R] Overdispersion in count data

2008-04-03 Thread Gavin Simpson
On Thu, 2008-04-03 at 01:24 +, David Winsemius wrote: > "Wade Wall" <[EMAIL PROTECTED]> wrote in > news:[EMAIL PROTECTED]: > > > Thanks for the recommendations, insights. I tried using glm.nb, but > > it didn't seem to like my data. I received the message (subscript) > > logical subscript t

Re: [R] Overdispersion in count data

2008-04-03 Thread Prof Brian Ripley
On Thu, 3 Apr 2008, Wade Wall wrote: > That is exactly how I am writing it. Glm works fine, but as I stated the > residual deviance is much greater (10x) than the degrees of freedom. I want > to take a look at using the negative binomial distribution, but I can't get > glm.nb to work. I get the

Re: [R] Overdispersion in count data

2008-04-03 Thread Michael Dewey
At 12:54 03/04/2008, Wade Wall wrote: >That is exactly how I am writing it. Glm works fine, but as I >stated the residual deviance is much greater (10x) than the degrees >of freedom. I want to take a look at using the negative binomial >distribution, but I can't get glm.nb to work. I get the m

Re: [R] Overdispersion in count data

2008-04-03 Thread Wade Wall
That is exactly how I am writing it. Glm works fine, but as I stated the residual deviance is much greater (10x) than the degrees of freedom. I want to take a look at using the negative binomial distribution, but I can't get glm.nb to work. I get the message Error: (subscript) logical subscript t

Re: [R] Overdispersion in count data

2008-04-03 Thread Michael Dewey
At 17:03 02/04/2008, Wade Wall wrote: >Hi all, > >I have count data (number of flowering individuals plus total number of >individuals) across 24 sites and 3 treatments (time since last burn). >Following recommendations in the R Book, I used a glm with the model y~ >burn, with y being two columns (

Re: [R] Overdispersion in count data

2008-04-02 Thread David Winsemius
"Wade Wall" <[EMAIL PROTECTED]> wrote in news:[EMAIL PROTECTED]: > Thanks for the recommendations, insights. I tried using glm.nb, but > it didn't seem to like my data. I received the message (subscript) > logical subscript too long. I am using the same dataframe as my > previous glm. Do you

Re: [R] Overdispersion in count data

2008-04-02 Thread Ben Bolker
Wade Wall gmail.com> writes: > > Thanks for the recommendations, insights. I tried using glm.nb, but it > didn't seem to like my data. I received the message (subscript) logical > subscript too long. I am using the same dataframe as my previous glm. Do > you know if I need to put the data in

Re: [R] Overdispersion in count data

2008-04-02 Thread Wade Wall
Thanks for the recommendations, insights. I tried using glm.nb, but it didn't seem to like my data. I received the message (subscript) logical subscript too long. I am using the same dataframe as my previous glm. Do you know if I need to put the data in a different format? Thanks, Wade On We

Re: [R] Overdispersion in count data

2008-04-02 Thread Achim Zeileis
On Wed, 2 Apr 2008, Gavin Simpson wrote: > On Wed, 2008-04-02 at 12:03 -0400, Wade Wall wrote: > > Hi all, > > > > I have count data (number of flowering individuals plus total number of > > individuals) across 24 sites and 3 treatments (time since last burn). > > Following recommendations in the

Re: [R] Overdispersion in count data

2008-04-02 Thread Gavin Simpson
On Wed, 2008-04-02 at 12:03 -0400, Wade Wall wrote: > Hi all, > > I have count data (number of flowering individuals plus total number of > individuals) across 24 sites and 3 treatments (time since last burn). > Following recommendations in the R Book, I used a glm with the model y~ > burn, with y

[R] Overdispersion in count data

2008-04-02 Thread Wade Wall
Hi all, I have count data (number of flowering individuals plus total number of individuals) across 24 sites and 3 treatments (time since last burn). Following recommendations in the R Book, I used a glm with the model y~ burn, with y being two columns (flowering, not flowering) and burn the time