Autori
Liseo, Brunero
Fortini, M.
Nuccitelli, A.
Scanu, M.

Titolo
Bayesian approaches to record linkage
Periodico
Università degli Studi di Roma "La Sapienza" - Dipartimento di Studi Geoeconomici, Linguistici, Statistici e Storici per l'Analisi regionale. Working papers
Anno: 2000 - Fascicolo: 15 - Pagina iniziale: 1 - Pagina finale: 30

Record linkage refers to the use of an algorithmic technique to match records from different data sets that correspond to the same statistical units. In this paper we propose a fully Bayesian approach to record linkage. We consider a classical measure of performance (the False Match Rate, see Belin et al. 1995) to construct a reasonable loss function and, according to the Bayesian paradigm, we look for the global action that minimises the expected posterior loss. Alternative to the formal decision theoretic analysis, we also propose a more ecletic approach which slightly deviates from the Bayesian road at the reward of a more flexible analysis




Esportazione dati in Refworks (solo per utenti abilitati)