Description: Actuaries can empower their understanding of policyholder behavior with third party data. Modeling policyholder behavior should start from developing knowledge of who they are, instead of actuaries assuming what a rational behavior should be.

Description: Matrix decomposition techniques are used to find main features of a dataset. These features can be used to describe the data or to find clusters of objects. The decompositions have problems when the matrices have missing data or are from different exposures. This article examines the issue and some papers dealing with it.