A note on the use of binomial distributions in CUB-type mixture models
In CUB-type mixture models, the binomial distribution is typically employed to model the deliberate choice or “feeling” component. It is shown that this specification is equivalent to choosing an adjacent categories model with specific constraints imposed on the systematic component. Embedding CUB-t...
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Article (Journal) |
| Lenguaje: | inglés |
| Publicado: |
October 2026
|
| In: |
Statistics & probability letters
Year: 2026, Volumen: 237, Pages: 1-5 |
| ISSN: | 1879-2103 |
| DOI: | 10.1016/j.spl.2026.110834 |
| Acceso en línea: | Verlag, kostenfrei, Volltext: https://doi.org/10.1016/j.spl.2026.110834 Verlag, kostenfrei, Volltext: https://www.sciencedirect.com/science/article/pii/S0167715226001987 
|
| Notas de Autor: | Gerhard Tutz, Moritz Berger, Ingrid Mauerer |
| Sumario: | In CUB-type mixture models, the binomial distribution is typically employed to model the deliberate choice or “feeling” component. It is shown that this specification is equivalent to choosing an adjacent categories model with specific constraints imposed on the systematic component. Embedding CUB-type models into the framework of adjacent categories models makes it easy to derive a straightforward interpretation of covariate effects. Moreover, a weaker motivation for the specification of the deliberate choice model is given. A real-world data application demonstrates the use of likelihood-ratio tests for model comparisons. |
|---|---|
| Notas: | Online verfügbar: 20. Mai 2026 Gesehen am 17.06.2026 |
| Descripción Física: | Online Resource |
| ISSN: | 1879-2103 |
| DOI: | 10.1016/j.spl.2026.110834 |