Roundtable: "The Limits of Models" Emanuel Derman, joining a distinguished panel of experts, advised, "Find the model that best approximates your world...so the more models the better. Be aware of when your model is wrong and the corrections that need to be made. Use variables that are intuitive and easily understood. Make sure the models are calibrated properly. Use financial concepts--not math concepts." No subject has been more central to this magazine than the debate about VAR modeling and other risk management techniques. Starting with the Group of 30 recommendations about risk in 1993, many people were already adapting VAR for use by corporations and hedge funds by 1995. We began a series of articles, roundtable discussions and columns on the topic. Some of the spirited debate we presented would prove prescient by the time Long-Term Capital Management went awry in 1998. Roundtable: "The Limits of VAR" "I find myself pleased with the quantitative developments of VAR. But let's take a look at the derivatives problems that have occurred. Barings was rogue trading. Orange County was rogue trading and leverage. Korea was leverage. Mexico was leverage and/or undemocratic capital markets. So if you focus on quantitative techniques, you miss virtually every major problem that systemically occurred," said Richard Sandor in a debate with risk experts that included Ron Dembo, Stan Jonas, William Margrabe and others. "Controlling Model Risk" In September 1998, Business Week wrote the following: "Long before the latest crisis, Derivatives Strategy ran as its June 1997 cover story an article entitled 'Controlling Model Risk.' It contained a jarringly prescient exposition of the problems lurking in multifaceted computer models." Plus "Taking the Stress Out of Stress Testing" "Hedging Efficiently with VAR" "Analyzing VAR" "The VAR Debate: Matching Objectives with the Application" "Estimating Volatility" "Worrying About Correlation" "Growing Pains of Monte Carlo Simulations" "Making VAR More Flexible" "Adapting Value at Risk"