10/05/2006

Market Timing and Asset Allocation

By Martin Gremm

Whenever an asset class experiences a correction or bear market, market timing enters the discussion. Understandably, we all would prefer to participate in the asset class only when it is going up. Pivot Point has long maintained that successful market timing is close to impossible to accomplish.

Whether market timing works remains controversial(see here for a survey). Hulbert Financial Digest is an independent organization that tracks market timers and other investment newsletters mostly with a stock market focus. Their findings suggest that successful market timing may be possible, but it is very difficult. Only a small fraction of strategies they track generate risk adjusted returns that surpass a simple buy and hold approach. In fact, the bulk of the top long term performers monitored by Hulbert are stock picking strategies similar to the portfolios offered by Pivot Point Advisors.

However, at least one mutual fund timing strategy tracked by Hulbert has generated superior risk adjusted returns for the last 15 years (Bob Brinker). At first blush, strong performance over such a long period of time is unlikely to be an accident. However, it turns out that Brinker's performance hinges on only three correct calls: a long signal in the early '90s, a cash position through the bear market post 2000, and a long position since 2003. Three trading decisions are not enough to evaluate the efficiency of a strategy. It is difficult to distinguish between somebody who got lucky three times and an approach that provides a real benefit.

Brinker's strategy, is supposed to be based on a mathematical model. Click here for an article comparing the performance of quantitative strategies to that of human experts in finance, wine tasting, medicine and other disciplines. Pivot Point Advisors uses mathematical models to select the stocks in our portfolios. The examples in this article support our view that a disciplined analytical approach has the potential to outperform the so-called experts.

This example of an apparently successful market timer has rekindled our interest in such strategies as part of an asset allocation model. Research is currently underway to explore whether an asset allocation based on a mathematical timing model has the potential to outperform a more static allocation strategy. We expect it will take at least six months to reach an answer and develop a strategy if initial results looks promising.

Are you interested in trying out various market timing and other investment strategies without risking capital? A number of websites allow you to track the performance of paper trades. Investing Experience has a day trading focus. For a site with an educational focus click here. One version of this simulator is designed as an educational tool for students. Another educational site can be found at Learning Investing and the link below is for a site that focuses on market timing strategies. These sites provide a framework for objectively testing the performance of investment programs, a very essential step in developing viable trading strategies.