The Variance Prize for papers published in Variance volume 10 has been awarded to Greg Taylor and Jing Xu for their paper “An Empirical Investigation of the Value of Claim Closure Count Information to Loss Reserving.” The prize was announced at the 2017 CAS Annual Meeting in Anaheim, California.
The winning paper tests whether loss reserving models that rely on claim count data can produce better forecasts (in the sense of being subject to lower prediction error) than the chain ladder model (which does not rely on counts). The authors find what they believe to be a compelling narrative. For their test data, the success of the chain ladder approach is limited; one or both of the models they test based on claim count data exhibit superior performance, in terms of prediction errors, the majority of the time.
The Variance Prize honors original thinking and research in property-casualty actuarial science. It is awarded to the author or authors of the best paper published in each volume year. To be eligible, a paper must show original research and the solution of advanced insurance problems.
For a number of years, Greg Taylor has been adjunct professor in the School of Risk and Actuarial Studies at The University of New South Wales, Sydney, Australia. Prior to his current post, he spent about 35 years as a consulting actuary. He is a Fellow of both the Institute and Faculty of Actuaries and the Institute of Actuaries of Australia. He holds doctorates in actuarial mathematics and theoretical physics. He has published three books on loss reserving, one of which is “Stochastic Loss Reserving Using Generalized Linear Models,” a CAS monograph cowritten with Gráinne McGuire.
Dr. Jing Xu received his Ph.D. in statistics from Macquarie University in Sydney, Australia, in 2014. His doctoral thesis focused on developing novel techniques in survival analysis. Other research projects have been related to modeling total loss reserves for general insurance companies. His research interests are in the areas of actuarial science, statistical methodology and applied statistics in medicine or health. Dr. Xu is currently a biostatistician at Singapore Clinical Research Institute.
The paper is published in Variance 10:1.