In several economic applications (e.g. marketing research, microsimulation models) there is the need to consider different data sources and to integrate the information coming from them. In this paper we show how integration problems can be managed by means of coherence for partial conditional probabilistic assessments. Coherence allows us to combine the knowledge coming from the different sources, included those (possibly) given from field experts, without necessarily assuming further hypothesis (as conditional independence). Moreover, inferences and decisions can be drawn taking in consideration also logical constraints among the variables. An example showing advantages and drawbacks of the proposed method is given.
Keywords. Coherent conditional probability, data fusion, statistical matching, inference.
The paper is availabe in the following formats:
Dipartimento Metodi e Modelli Matematici
Universita' La Sapienza
via Scarpa n. 16 Roma,