Beta-N-Mixture

This repository contains the source code necessary for estimating abundance using the Beta and Normal N-mixture models. The specification of both multispecies models are written in BUGS language and we also provide R functions for abundance estimation using single species N-mixture models. The data to the three steps of the Beta N-mixture validation are separated in different .RData files. The data sets for the 1500 simulations with hi, mid and low detection probability are saved in the bias.RData. The 500 data sets simulated under the complicated model used to compare the Beta and Normal N-mixture model along with the "lambda" and "p" used in each simulation are saved under the comparison.RData. The real count data from the point counts performed in central Colombia are saved in the file real.RData. The entire code is saved in the Gomez_et_al_code.R and has been annotated to be self explanantory. All of the analysis of the paper should be replicated from the source code. The only step for which we did not save the simulated data was the bias estimation of the single species N-mixture model because of the large number of simulations performed. Using the code and function provided, however, the reader should be able to reproduce the simulations and the bias estimation. https://doi.org/10.5281/zenodo.827266