Last month, Katz presented his experience developing a trading system using a neural network-based approach. This month, he delves into a rule-based approach to trading."Previously, I showed that attempts to model the markets by applying sophisticated technology such as neural networks in a simplistic manner does not work well. Present-day attempts to apply the neural-net novice approach fail because the market inefficiencies discovered by the simplistic application of sophisticated technologies have already been "traded away" by speculators who used the technology to discover and trade the inefficiencies. "
The rule-based approach may also be a "mind-based" one, relying on the individual constructing the system. Once observations have been made, these ideas must be formalized into computer programming instructions.
The Nature of the System
"In examining charts, I have observed that some markets such as the Standard & Poor's 500 index and bonds frequently move in thrusts and retracements, even in an intraday time frame. If the overall market direction is bullish, often the market will open, inch down to form a low, make a larger thrust upward to form a high, pull back to a retracement level, and then make another upward thrust. In a bearish market, of course, the pattern is reversed, with the thrusts down and the retracements up."Figure 2: TYPICAL TRADING BEHAVIOR ON 15-MINUTE BONDS. The trading behavior of this system being developed is illustrated here. Many entries occur precisely as illustrated at the bottoms of retracements - that is, at exact pivot lows. The profit targets are also good, with the exits taking place very close to the highest highs of the day. On trades that fail, the stop gets the market participant out very quickly. "My first step in developing this system was to write a set of rules that would express my concept of the thrust-retracement pattern. My main concern was to make sure there would be a reasonable match between the rule-generated identifications of the thrust retracement pattern and those generated subjectively."