This research presents a new probabilistic approach for the analysis of preference data when dealing with paired comparisons. The combination of the extended log-linear Bradley-Terry model with the regression trunk methodology generates a new model that allows finding interaction effects between subject-specific covariates. By fitting Poisson regressions to find the best split points and applying the final pruning procedure, the result is a small regression tree (so-called regression trunk). It represents a compromise between an easier interpretation of higher-order interaction effects and an efficient partition of individuals according to their preference scales.

Discovering Interaction Effects Between Subject-Specific Covariates: A New Probabilistic Approach For Preference Data / Alessio, Baldassarre; Claudio, Conversano; D'Ambrosio, Antonio; Mark De Rooij, ; Elise, Dusseldorp. - (2020), pp. 1166-1170. (Intervento presentato al convegno sis 2020).

Discovering Interaction Effects Between Subject-Specific Covariates: A New Probabilistic Approach For Preference Data.

Antonio D’Ambrosio;
2020

Abstract

This research presents a new probabilistic approach for the analysis of preference data when dealing with paired comparisons. The combination of the extended log-linear Bradley-Terry model with the regression trunk methodology generates a new model that allows finding interaction effects between subject-specific covariates. By fitting Poisson regressions to find the best split points and applying the final pruning procedure, the result is a small regression tree (so-called regression trunk). It represents a compromise between an easier interpretation of higher-order interaction effects and an efficient partition of individuals according to their preference scales.
2020
9788891910776
Discovering Interaction Effects Between Subject-Specific Covariates: A New Probabilistic Approach For Preference Data / Alessio, Baldassarre; Claudio, Conversano; D'Ambrosio, Antonio; Mark De Rooij, ; Elise, Dusseldorp. - (2020), pp. 1166-1170. (Intervento presentato al convegno sis 2020).
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11588/820370
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact