Home2018-09-29T03:53:13+00:00

Risk Analysis, Management and Modeling for the Insurance and Financial Industries

Research Papers
Discussion Papers
Magazine Articles
Alphabetical
Unpublished
Presentations
Research Ideas

About Gary

Gary retired from the insurance industry after a 43-year career working at two primary companies, a reinsurer, a broker, a pool, and a rating bureau. He is now an adjunct professor for research at the University of New South Wales and teaches an actuarial graduate course on quantitative risk management at Columbia University, with an emphasis on building economic scenario generators and economic capital models.

His career focus was using advancements in computing and statistical methodology to improve actuarial models. The IT revolution made the last 50 years a good time for this, with no end in sight. His current research interests include using Bayesian shrinkage in MCMC for pricing, reserving, and mortality models, capital needs and allocation, and economic scenario generators.

Gary’s Bio | Gary on Google Scholar | Email Gary

Research Ideas

The Quiet Revolution in Statistics

A somewhat dramatic shift in statistical methodology is underfoot. For regression-type models the basic result is that Bayesian shrinkage produces lower fitting and prediction errors than does MLE. There have been several larger shifts going on that drive this: Bayesians are becoming less Bayesian and frequentists are becoming more Bayesian. A meeting point [...]

Paper Getting Read

Gary’s paper, "REGULARIZED REGRESSION FOR RESERVING AND MORTALITY MODELS", was recently listed on six SSRN's Top Ten download lists: ERN: Microeconometric Studies of Health Markets (Topic) Top Ten, ERN: Model Construction & Estimation (Topic) Top Ten, ERN: Uncertainty & Risk Modeling (Topic) Top Ten, Econometric Modeling: Microeconometric Studies of Health, Education, & Housing [...]

Capital Needs of an Insurance Company

There are two historical trends on capital needs and risk, tracing back to the work of two Italian researchers in the late 1950s. Franco Modigliani (with Merton Miller) argued that risk management that costs something in expected earnings is not worthwhile. One of their assumptions was that a firm that gets into financial [...]

Load More Posts

Gary in Singapore

Regularized Regression for Reserving and Mortality Models

Paper Summary and Video