Chapter 6: History

Figure 6.9 Dow Jones Industrial Average with 21-day moving average and 4 percent bands. Many a market timing system was built on percentage bands drawn around the Dow Jones Industrial Average.

signals for stocks involved volume oscillators. This work was indicative of a broad trend to systematize decision making using technical analysis.

Up to this point, all approaches to bands and envelopes were symmetrical. In the early 1980s, Marc Chaikin working with Bob Brogan produced the first fully adaptive bands. Called Bomar Bands (BOb and MARc), these were trading bands that contained 85 percent of the price action over the past year, as you can see in Table 6.3. The importance of this achievement cannot be

Table 6.3 Bomar Band Formulas

Bomar Bands

Upper band contains 85 percent of data above the average for the past 250 periods

Middle band - 21-day moving average

Lower band contains 85 percent of data beneath the average for the past 250 periods

Part II: The Basics

AMZN AMAZON COMSTK

Hi: 13.34 Lo: 18.50 CI : 10.50

oi a /oi

15%: 19.32

%: 1 .

18.50 -1.44

Dail

10 Sep

Figure 6.10 Bomar Bands, (source: Instinets Research and Analytics.)

overstated. In strong uptrends the upper Bomar Band would widen appropriately while the lower Bomar Band contracted. Volatile stocks had wide bands, whereas stable stocks had narrow bands. In downtrends the lower band expanded and the upper contracted (Figure 6.10). Thus, Bomar Bands not only broke with the idea that bands should be symmetrical but evolved over time to suit the price structure.

The major benefit conferred by Bomar Bands was that analysts were no longer forced to provide their own guesses about what the proper values for the bands were. Instead, they were free to focus on decision making and let their PCs set the bandwidth for them. Unfortunately, Bomar Bands were extremely computationally intensive for their time, and to this day are not readily available beyond Instinef s research and analytics (R&A) platform. Thus they have not achieved the broad acceptance they deserve.

The late Jim Yates of DYR Associates, working in the late 1970s and early 1980s, provided an important insight using implied volatility from the options market. He derived a method of determining whether a security was overbought or oversold in relation to market expectations. Mr. Yates showed that expectations of volatility could be used to create a framework within

Chapter 6: History

which one could make rational decisions regarding stocks and or options. This framework consisted of six zones and mapped out the appropriate option strategies for each zone (Figure 6.11). This became his Options Strategy Spectrum, which remains a useful tool to this day under the care of his son Bill.

What Jim did was to create zones (bands) based on the implied volatility of options. Then he used those bands to determine what strategies were most appropriate given market conditions. It was a brilliant insight that foreshadowed much of the work I was to do.

As the 1980s dawned, I was active in the options markets, the key to which is an understanding of volatility in its many forms, though I was not fortunate enough to have met Jim Yates yet. I also was very interested in technical analysis and specifically in trading bands. It occurred to me that the key to the proper bandwidth was volatility. So I embarked on a testing program that examined various measures of volatility as a method of setting bandwidth. It became readily apparent that the standard deviation calculation provided a superior result. This is primarily due to the squaring of the deviations from the average in the calculation.6

«S ¹KUH JUN M *U0 MP OCT NOV MC J*N ItO 1*

Figure 6.11 Implied risk indicator, (source: The Options Strategy Spectrum by James Yates, Homewood, 111.: Dow Jones-Irwin, 1987.)