AICcmodavg: R Package for Model Selection and Multimodel Inference

An R package providing functions for model selection and multimodel inference based on Akaike's information criterion (AIC, AICc) and their quasi-likelihood counterparts (QAIC, QAICc). The package, authored by Marc J. Mazerolle, implements model averaging for parameters or predictions, includes diagnostics for certain model types, and supports Bayesian models from 'bugs', 'rjags', and 'jagsUI' classes. It also allows for model selection using BIC and can format results to LaTeX.

Use Cases

• Compare statistical models based on AIC, AICc, QAIC, or BIC values as described in the package functions.
• Perform model averaging for parameter estimates or predictions using the classic or shrinkage methods implemented.
• Calculate goodness-of-fit statistics for 'unmarkedFit' class models estimating demographic parameters.
• Create model selection tables for Bayesian models from 'bugs', 'rjags', or 'jagsUI' classes.

Strengths

• Implements a suite of established model selection criteria including AIC, AICc, QAIC, QAICc, and BIC.
• Includes specialized diagnostics for 'unmarkedFit' classes used in ecological studies with imperfect detection.
• Offers compatibility with multiple Bayesian modeling frameworks ('bugs', 'rjags', 'jagsUI').

Limitations

• Row count and data scale are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Marc J. Mazerolle

Collection Method

Software package development for statistical analysis in R.