Individual Claim Development Models and Detailed Actuarial Reserves in Property-Casualty Insurance

Author: Chris Gross – Chief Executive Officer at Gross Consulting


Individual Claim Development Models and Detailed Actuarial Reserves in Property-Casualty Insurance


Actuarial reserving techniques using aggregated triangle data are ubiquitous in the property casualty insurance industry. By instead starting with the modeling of individual claim behavior using predictive modeling techniques and a modeling framework that describes the full life cycle of a claim, there are numerous benefits including greater reliability of reserve estimates, faster recognition of underlying mix changes, and avoidance of problems in pricing due to differences in development. Component development and emergence models used in conjunction with simulation of currently outstanding claims and simulation of claims still yet to be reported form an alternative framework for generating estimates of reserve need. Algorithmic case reserves at the claim level and algorithmic IBNR estimates at the policy level, actuarially determined and designed to be unbiased, provide valuable information for downstream analyses, a bridge to the generally accepted triangle reserving paradigm, and a means for demonstration of reliability for actuarial purposes.


Chris Gross Co-Authors Variance Journal Paper

Validation of minimum bias rate factors

Released in the December, 2018 issue of Variance Journal, Christopher Gross and Jonathan Evans co-authored a paper entitled Minimum Bias, Generalized Linear Models, and Credibility in the Context of Predictive Modeling.

Abstract: When predictive performance testing, rather than testing model assumptions, is used for validation, the need for detailed model specification is greatly reduced. Minimum bias models trade some degree of statistical independence in data points in exchange for statistically much more tame distributions underlying individual data points. A combination of multiplicative minimum bias and credibility methods for predictively modeling losses (pure premiums, claim counts, average severity, etc.) based on explanatory risk characteristics is defined. Advantages of this model include grounding in long-standing and conceptually lucid methods with minimal assumptions. An empirical case study is presented with comparisons between multiplicative minimum bias and a typical generalized linear model (GLM). Comparison is also made with methods of incorporating credibility into a GLM.

Download the full study directly from Variance Journal here.