We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Methodology

Title: Mixture composite regression models with multi-type feature selection

Abstract: The aim of this paper is to present a mixture composite regression model for claim severity modelling. Claim severity modelling poses several challenges such as multimodality, heavy-tailedness and systematic effects in data. We tackle this modelling problem by studying a mixture composite regression model for simultaneous modeling of attritional and large claims, and for considering systematic effects in both the mixture components as well as the mixing probabilities. For model fitting, we present a group-fused regularization approach that allows us for selecting the explanatory variables which significantly impact the mixing probabilities and the different mixture components, respectively. We develop an asymptotic theory for this regularized estimation approach, and fitting is performed using a novel Generalized Expectation-Maximization algorithm. We exemplify our approach on real motor insurance data set.
Subjects: Methodology (stat.ME); Econometrics (econ.EM); Applications (stat.AP)
Cite as: arXiv:2103.07200 [stat.ME]
  (or arXiv:2103.07200v2 [stat.ME] for this version)

Submission history

From: Tsz Chai Fung [view email]
[v1] Fri, 12 Mar 2021 10:48:38 GMT (1453kb,D)
[v2] Fri, 30 Jul 2021 15:14:17 GMT (1488kb,D)

Link back to: arXiv, form interface, contact.