Effect Displays in R for Multinomial and Proportional-Odds Logit Models: Extensions to the effects Package

Abstract

Based on recent work by Fox and Andersen (2006), this paper describes substantial extensions to the effects package for R to construct effect displays for multinomial and proportional-odds logit models. The package previously was limited to linear and generalized linear models. Effect displays are tabular and graphical representations of terms — typically high-order terms — in a statistical model. For polytomous logit models, effect displays depict fitted category probabilities under the model, and can include point-wise confidence envelopes for the effects. The construction of effect displays by functions in the effects package is essentially automatic. The package provides several kinds of displays for polytomous logit models.