UB models are a class of mixture distributions for analyzing ordinal responses in the form of ratings or marginal ranking that designs the decision process as the combination of two main components: a personal feeling and an inherent uncertainty. This methodology is receiving an increasing interest due to successful applications to the analysis of evaluations, opinions and perceptions, in various fields of research, and thanks to advantageous graphical interpretation of results. Currently the estimation of CUB models can be run by the R library ’CUB’ ver. 1.1.3. In this paper a gretl package for CUB models is presented, all the estimation examples and graphs proposed in the R vignette are replicated.
CUB for gretl / Simone, Rosaria; DI IORIO, Francesca; Lucchetti, Riccardo. - Scuola delle Scienze Umane e Sociali. Quaderni:12(2019), pp. 147-166. (Intervento presentato al convegno 6th Gretl Conference tenutosi a Dipartimento di Scienze Politiche, Università di Napoli Federico II nel 13-14 Giugno 2019) [10.6093/978-88-6887-057-7].
CUB for gretl
Rosaria Simone;Francesca Di Iorio
;Riccardo Lucchetti
2019
Abstract
UB models are a class of mixture distributions for analyzing ordinal responses in the form of ratings or marginal ranking that designs the decision process as the combination of two main components: a personal feeling and an inherent uncertainty. This methodology is receiving an increasing interest due to successful applications to the analysis of evaluations, opinions and perceptions, in various fields of research, and thanks to advantageous graphical interpretation of results. Currently the estimation of CUB models can be run by the R library ’CUB’ ver. 1.1.3. In this paper a gretl package for CUB models is presented, all the estimation examples and graphs proposed in the R vignette are replicated.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
