Given the fitted parameter values for a log-linear model, compute an information criterion.

1 |

`predictors` |
A character vector of predictors of the form "c1", "c2" for main effects, or "c12" for an interaction. The predictors to be used in a log-linear model. For example, "c1", "c2" for main effects, or "c12" for an interaction. |

`ddat` |
A data frame that is the design matrix for a log-linear model. |

`ic` |
The information criterion to be computed. Currently the AIC, AICc, BIC, BICpi are implemented. |

`beta` |
The vector of log-linear coefficients that were previously estimated. |

Computes the conditional multinomial likelihood and uses it to compute the specified information criterion

The value of the information criterion

Zach Kurtz

Thesis of Zach Kurtz (2014), Carnegie Mellon University, Statistics

Anderson DR and Burnham KP (1999). "Understanding information criteria
for selection among capture-recapture or ring recovery models." *Bird
Study*, **46**(S1), pp. S14-S21.

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.