[R-sig-eco] Bayesian analysis of canopy cover/percent data

Hi everyone, I'd like to estimate the means and sd's of canopy cover for 13 sites (9 are birds and 4 are potential habitats). I'm using a beta distribution to account for the fact that the data are bound by 0 and 1. For each sample, I measured the proportion of points that have some sort of vegeta

[R-sig-eco] stochastic population models

Hi everyone, I would like to create stochastic population models for an undergraduate course. The goal would be to have students run models, record, results, change parameters, and make inferences on changing the effects. I understand how to draw from distributions, I hit a knowledge wall with l

[R-sig-eco] coding question

Hi everyone, I'd like to compare vegetation in forest fragments of various sizes (1-, 10, and 100-ha) with vegetation where nine species of birds forage. I consider the bird samples to be independent, even within the same species. I have, however, several samples within each fragment. I'd like to

[R-sig-eco] R2OpenBUGS help

Hi everyone, I recently did an analysis using WinBUGS to describe the vegetation associated with a number of South American birds (Journal of Field Ornithology 84:1-12). Now I would like to do the same analysis but compare the vegetation where birds forage with random sites in forest fragments wh

[R-sig-eco] repeated measures with multiple comparisons

Hi everyone, I'm helping some colleagues with analyzing behavioral data. The experiment was to repeatedly remove seeds that animals have cached and see what they do after a number of trials (control + 9). Here's some of the code for the repeated measures: squirrel <- read.csv("h:/temp/s

Re: [R-sig-eco] papers about using MCMC in ecological study

Marc has two books out: 1. Introduction to WinBUGS for Ecologists: Bayesian approach to regression, ANOVA, mixed models and related analyses 2. Bayesian Population Analysis using WinBUGS: A hierarchical perspective I have the first and I probably reference it once a week. I'm not a statisticia

[R-sig-eco] testing for synchrony

Hi everyone, I've been asked a few times about how to test for synchrony of ecological events. In one case, a researcher has more than a decade of acorn mast data for several species of oaks and wants to know if different oaks have synchronous mast years when sympatric. The other case, a researc

Re: [R-sig-eco] zero-truncated Poisson

Thanks Ben and Peter for your thoughtful replies. This gives me quite a bit to chew on and digest but I think I can analyze these data. -Original Message- From: Ben Bolker [mailto:bbol...@gmail.com] Sent: Sunday, April 10, 2011 10:38 AM To: Peter Solymos Cc: Stratford, Jeffrey; r-sig

Re: [R-sig-eco] zero-truncated Poisson

893878, 2.28446939, 4.56893878, 2.28446939, 4.56893878 )), .Names = c("year", "size", "distance", "taken", "mass"), class = "data.frame", row.names = c(NA, -550L)) -Original Message- From: Ben Bolker [mailto:bbol...@gmail.com] Sent: Friday

[R-sig-eco] zero-truncated Poisson

Hi everyone, I'm attempting to run a zero-truncated Poisson using VGAM and I run into this error: Error in if ((temp <- sum(wz[, 1:M, drop = FALSE] < wzepsilon))) warning(paste(temp, : argument is not interpretable as logical Here's the code I'm using blja <- read.csv("g:\\blja\\blja.csv",

Re: [R-sig-eco] Negative binomial

Oi Joao, >From my understanding of the Poisson, as the Poisson mean (lambda) increases, >the distribution converges with the mean. I believe there are also issues >with the intercept since counts cannot be negative. If the variance is >excessive (var/df >> 1) then switch to negative binomial.