Abund.md
In shinyrecap: 'Shiny' User Interface for Multiple Source Capture Recapture Models

Model Comparison

Each row summaries the results for a log linear model type.

Model Types: M0: This model assumes equal capture probability and homogeneity. Mt: This model assumes homogeneity. Mh: This model assumes equal capture probability. M0: This model assumes neither.

Heterogeneity Types: Normal: The log odds of capture follows a Normal distribution. Darrosh: The log odds of capture among those who were not captured follows a Normal distribution. Poisson: The log odds of capture among those who were not captured follows a Poisson distribution. Gamma: The log odds of capture among those who were not captured follows a Gamma distribution.

Columns are defined as follows: Population Size: This is the population size as estimated by each model. strerr: This is the standard error of the population size. AIC: This is the Akaike Information Criterion, and is a good measure to use to select which model to report. Lower is better. BIC: This is the Bayesian Information Criterion, and is also a good measure to use to select which model to report. It generally favors simpler model (i.e. ones with more assumptions) than the AIC. Lower is better.