 |
March 1997
Anyone developing a model to time the stock market will discover that there are important differences between today's markets and the markets 20 years ago. The most obvious: Most indices have skyrocketed! This creates a problem in that a 20-point change in the Dow Jones Industrial Average (DJIA) today is nothing, but it was a major event in the 1970s. The solution is to convert your data to percentages to reflect this dramatic change in scales.Of course, this isn't the only dilemma that you need to consider, and that leads me to this month's feature, "A New Utility Average Stock Market System" by STOCKS & COMMODITIES Contributing Editor Dennis Meyers, a followup to a piece in our August 1995 issue. In that issue, Meyers detailed a system for market timing the Standard & Poor's 500 index; the article this time follows the same line of reasoning, except he incorporates changes to address certain concerns. For example, Meyers reasons that weekly data can leave a trader exposed to large intraweek fluctuations, so the new system he details uses daily data instead of weekly. In addition, the data is preprocessed to be equivalent to percentages to account for the changes in scale.
These considerations, as well as other issues, could occur to anyone working on market timing models, and that's why I always recommend Meyers's work for anyone developing timing models. After all, his background includes a doctorate in applied mathematics in engineering and a membership on the Chicago Board Options Exchange (CBOE); with more than 25 years' experience in the markets, he's seen a lot. His articles hit on issues that come up sooner or later as you work on your own systems. You may not be interested in a timing model for the stock market, but as you develop your own trading plan, there's a good chance that he's dealt with problems you'll run into.
This time, Meyers uses the Dow Jones Utility Average as a key indicator for a number of reasons, one of which is this index has been a good proxy for interest rates, and as seasoned investors know, the direction of interest rates is a leading indicator for the stock market. The bond market, probably the key indicator of interest rates, has its own characteristics. And what might those be? To provide insight on that subject, check out this month's interview, this time with technical analyst and Salomon Brothers vice president Alex Saitta.
Saitta's expertise is interest rates, though he discusses a number of topics, including his research into common technical observations, such as intermarket relationships; but -- and this is an important "but" -- he supports his observations with research. In fact, Saitta is adamant about testing technical methods. He avers that if a method cannot stand up to critical evaluation, then he believes that it has no value. Fortunately for us, Saitta has additional articles in mind to explain his findings, and I look forward to presenting them to you.
One final point. Even though our authors are providing more and more supportive documentation, you shouldn't just blindly follow the system in question. You still need to sit down and look over the data, really look it over, test it out on your own, verify that the method or model is really what you're looking for. This, you'll find, makes all the difference for you between loss and profit. And isn't that why we're in this business?
