klue - Hybrid Machine Learning and Random-Utility Workflow for Latent Class Multinomial Logit Model Specification

Implements the three-step workflow from Frings (2026, working paper) for specifying latent class multinomial logit (LCMNL) models. The maximum-likelihood multi-start is initialised from six clusterings of respondents' revealed-preference signatures (k-means, Gaussian mixture, hierarchical clustering with Ward, complete, and average linkage, and partitioning around medoids); LCMNL is estimated across a user-specified range of class counts; and a mixed multinomial logit (MMNL) benchmark is reported alongside BIC, AIC, ICL, and a classification-entropy diagnostic. Accepts long- or wide-format discrete-choice data with optional availability columns. Validated against five public reference datasets (Vittel, Apollo mode and route choice, Electricity, Swissmetro). Wraps the 'apollo' package for maximum-likelihood estimation.