Re: [R-sig-eco] Logistic regression with repeated measures ?

I meant to write "logit transformation” not “logic”. I haven’t read too much about beta regression and the article I posted doesn’t mention it, but Peter’s advice sounds good too. Mollie Mollie Brooks, PhD Postdoctoral Researcher, Population Ecology Research Group ht

Re: [R-sig-eco] Logistic regression with repeated measures ?

Hi, Transforming the proportions and doing a linear regression is probably the best thing to do in this case. Some would recommend the arcsine transformation in this case, but a quick search of “arcsine logit proportion transform” gave this article that recommends the logic transformation The

Re: [R-sig-eco] Logistic regression with repeated measures ?

So you want like a dose-response relationship using proportions of time as response if I understand correctly. And you are worried that some overdispersion might be present. The beta regression deals with continuous proportions as outcome and you can model overdispersion (even as a function of cova

Re: [R-sig-eco] Logistic regression with repeated measures ?

Hello Chris and Peter,  Thank you very much for your quick answers. But as I pointed out, this study really is not a resource selection project (sorry I did not provide more background). We are not really interested in knowing why the birds are going to a specific habitat rather than an other

Re: [R-sig-eco] Logistic regression with repeated measures ?

Chris, It is not random effect strictly speaking, but something like that. The problem is this: RSF models are often constructed as mixed models with random intercept. But it is known that the intercept is a function of the other parameters and the available (background) distribution. So a random

Re: [R-sig-eco] Logistic regression with repeated measures ?

Hi Peter, Does it have the ability to fit random effects? Or some other way to address the pseudoreplication expected in RSF studies using GPS fix data with little time between fixes ? (Just had a quick look at the rspf package and I couldn't see any) Chris Howden B.Sc. (Hons) GStat. Founding P

Re: [R-sig-eco] Logistic regression with repeated measures ?

Marie, Your problem and data seems to me a resource selection problem with matched use-availability design. Estimating procedure for that design is discussed in Lele and Keim (2006, Ecology 87:3021--3028) and implemented in the ResourceSelection package: rspf function, see description of argument

Re: [R-sig-eco] Logistic regression with repeated measures ?

Hi Marieline, I would consider using the raw data so it is binary, and use a random intercept to account for the different sampling intensities for each bird. If U have summarised the data so it is 1 score for each bird which is a % I can’t see how U can account for repeated measures since U don

[R-sig-eco] Logistic regression with repeated measures ?

Dear list, I am a bit new to logistic regressions. I am working with GPS data from GPS-tracked birds. My objective is to investigate whether various covariates influence the probabilty of visiting specific habitats. Each bird has visited many habitats during the course of its GPS tracking.  He

[R-sig-eco] Modeling overdispersed count data

Bayesians commonly use the log normal to account for overdispersion. I learnt the trick from Ben Bolker. His book may be a citable source. Philip Dixon ___ R-sig-ecology mailing list R-sig-ecology@r-project.org https://stat.ethz.ch/mailman/listinfo/r

[R-sig-eco] Observation-level random effects to account for overdispersion

All, I am a real fan of using observation-level random effects to account for overdispersion with count data. Using the default log link, an observation-level normally-distributed random effect models the data as a convolution of a Poisson and log-normal distribution. The negative binomial is

Re: [R-sig-eco] Dealing with overdispersion in mixed model with count data

I believe Elston et al. 2001 (link) is usually cited for using observation-level random effects, but is appears to be treated skeptically by some (link