Collecting insurance fraud samples is costly and if performed manually is very time consuming. This issue suggests usage of unsupervised models. One of the accurate methods in this regards is Spectral Ranking of Anomalies (SRA) that is shown to work better than other methods for auto insurance fraud detection specifically. However, this approach is not scalable to large samples and is not appro...

Guido Calabresi’s book, THE COSTS OF ACCIDENTS, gives two contrasting examples which together prove that efficient decisions about car safety devices must be guided by individual costs. Part I of this essay reviews Calabresi’s demonstration that, if charged accurate insurance costs, Taney would invest in improved brakes and Marshall would not. To show why this demonstration fails to inform deba...

This paper generalizes Poisson-Multifractal for correlated time series of count data. We show that the model has useful properties; it captures long-term time dependence and exible dependence between types of count. Based on real data, the correlated multifractal model is used to model the number of claims of two separate coverages in automobile insurance. Smoothed values of the underlying proc...

Abstract The fact that a large proportion of insurance policyholders make no claims during one-year period highlights the importance zero-inflated count models when analyzing frequency claims. There is vast literature focused on univariate case models, while work in area multivariate considerably less advanced. Given companies write multiple lines business, where claim counts these business are...