 |
April 1997
The STOCKS & COMMODITIES interview this month is with money manager John Bollinger, perhaps best known for Bollinger bands, the popular technical indicator that bears his name. Not content to rest on his laurels with that, he has continued to research many facets of technical analysis, and his work has led him to some interesting conclusions; in fact, his points have me thinking about similar issues. The key one for me, which he discusses in our interview, is the use of a crisp value for technically based traders and investors. What are we talking about, lettuce? Of course not. No, a crisp value is a term used to distinguish between simple binary terms -- say, off and on, or in trading terms, in the market and out of the market. A simple example would be, you're invested if the Dow Jones Industrial Average (DJIA) moves above its 200-day moving average, and if the DJIA drops below the moving average, you're not.Bollinger points out that such a viewpoint isn't realistic; going from a fully invested position in the stock market one day and being completely on the sidelines the next is not a true representation of what traders do. That's a mindset more appropriate for the trader with a time horizon of perhaps just a few days. In such a situation, you do need to use crisp values, or price levels upon which to act. Investors, on the other hand, are oriented toward maintaining positions in concert with long-term trends, and long-term trends don't usually reverse on a dime. More important, the technical conditions are not going to match precisely for each market top, nor will the exact same technical conditions occur for each market bottom. Yet there can be a similarity of technical conditions.
To deal with this issue, Bollinger has been exploring the use of fuzzy logic, an approach that makes use of a range of values, compared with a single value triggering your decision. For example, say you use a sentiment indicator. It makes sense that the sentiment indicator will not hit the same level at each market reversal, and so you need to incorporate a method that permits some flexibility in the application of this indicator; after all, the market is dynamic, not rigid, and you need to follow suit. Bollinger's been working on this very problem, and after reading our interview, I'm sure that you'll be giving it some thought as well.
From time to time, I receive requests for more articles on indicators that are available in software packages. People would like to know more about how to use the tools they already have. Toward that end, this month's feature article, "The moving average convergence/divergence histogram," fits the bill. Author Mark Vakkur applies the MACD indicator as a trading system to a number of different markets, including market indices, mutual funds and stocks. Vakkur supplies numerous tables to support his conclusions. As a result, in Vakkur's piece, you have not only a system to consider but a review of historical performance too.
I can't stress how important it is to make similar efforts developing your own trading systems. The effort will pay off, I assure you, and in many different forms, ranging from knowing what to expect in the future to having a benchmark with which to gauge your performance. It's a lot easier to figure out where you're going if you have some guideposts to tell you that you're headed in the right direction, right?
