GMLTM - Generalized Multicomponent Latent Trait Model for Diagnosis

Provides Bayesian estimation of Item Response Theory models that decompose item difficulty into cognitive operations or rules. Implements the Linear Logistic Test Model (LLTM; Fischer (1973) <doi:10.1016/0001-6918(73)90003-6>), the Multicomponent Latent Trait Model for Diagnosis (MLTM-D; Embretson and Yang (2013) <doi:10.1007/s11336-012-9296-y>), and the Generalized Multicomponent Latent Trait Model for Diagnosis (GMLTM-D; Ramirez et al. (2024) <doi:10.3390/jintelligence12070067>), including a variant with correlated latent components. All models are estimated via Hamiltonian Monte Carlo using 'Stan' through the 'rstan' interface. Includes tools for prior predictive checks (Gelman et al., 2020), model validation, conditional reliability estimation, examinee mastery classification following Embretson (2019) <doi:10.1007/978-3-030-05584-4_9>, and individual diagnostic reports at the rule and component level. Supports user-defined prior distributions for all model parameters.