Fraud detection by a multinomial model. Separating honesty from unobserved fraud. A Monte Carlo simulation study (developed)

With the problem of detecting tax evasion in mind, we investigate how to identify items, e.g. individuals or companies, that are wrongly classified as honest. Normally, we observe two groups of items, labeled fradulent and honest, but suspect that many of the observationally honest items are, in fact, fraudulent. The items observed as honest are therefore divided into two unobserved groups honestH, representing the truly honest, and honestF, representing the items that are observed as honest, but that are actually fraudulent. By using a multinomial logit model and assuming commonality between the observed fraudulent and the unobserved honestF, Caudill, Ayuso, and Guillén (2005) present a method that uses the EM-algorithm to separate them. By means of a Monte Carlo study, we investigate how well the method performs, and under what circumstances. We then compare it to other standard methods.

Joint with; Jonas Andersson, and Aija Polakova.