Taylor and Xu Win 2016 Variance Prize
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 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 and 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.
Greg Taylor has been Adjunct Professor in the School of Risk and Actuarial Studies at The University of New South Wales, Sydney, Australia for a number of years. Prior to this, he spent about 35 years as a consulting actuary. He is a Fellow of the Institute and Faculty of Actuaries, and of the Institute of Actuaries of Australia. He holds a doctorate in actuarial mathematics, and a doctorate in theoretical physics. He has published three books on loss reserving, one of them a (joint work) CAS monograph.
Dr. Jing Xu received his PhD in Statistics from Macquarie University, Australia, in 2014. His PhD 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 area of actuarial science, statistical methodology and applied statistics in medicine or health. Dr. Xu is currently a biostatistician at Singapore Clinical Research Institute, Singapore.
The winning paper was published in Variance volume 10, number 1.