In the preceding section in this series, we look at how one company isolated parts of Elliott Wave patterns and helped the trader identify them in both end-of-day and real-time trading situations. In this section, we talk to an experienced trade systems designer who has researched the challenge of implementing Elliott Wave theory by computer since the mid-1990s.
Murray Ruggiero is no stranger to trading system users. He is the author of a number of books on the topic - including "Cybernetic Trading Strategies: Developing A Profitable Trading Strategy With State-Of-The-Art Technologies" (1997) and "Traders' Secrets Psychological & Technical Analysis: Real People Becoming Successful Traders" (1999), the Inside Advantage newsletter, as well as more than 70 articles in various trading publications. His work is referenced in books by such prominent authors as Larry Williams, John Murphy and Perry Kaufman.
In his book "Trading Systems And Methods" (1998), trading system specialist Perry Kaufman presents four of Ruggiero's suggestions for trading Elliott by machine:
1) Enter wave 3 in the direction of the trend.
2) Stay out of market during wave 4.
3) Enter wave 5.
4) Take countertrend ABC at top of wave 5.
Kaufman also says: "When a wave appears in two time frames such as both daily and weekly charts, the likelihood of the success of this formation increases." Without some sort of confirmation, the risk of being on the wrong side of the trade increases.
Accuracy is Not Key?
The problem with trading Elliott concepts by computer, Ruggiero believes, is that the designer must reduce the highly subjective aspect of the theory into quantifiable, specific components. The goal is to find those areas of the theory that work best and then tell the computer how to find them for you.
To Ruggiero, the key is not in trying to "teach" the machine to count Elliott Waves accurately because, like Robert Prechter, Ruggiero still believes that it takes a degree of human intervention to apply the highly complex aspects of Elliott Wave interpretation. This need for human involvement is due to the fact that Elliott Wave has been traditionally used in longer-term forecasting.
But traders are more interested in much shorter time frames, and it makes sense that a system that is to trade intraday has a different focus than a system looking for a target that is weeks, months or years away.
"There is a difference between today's count and the true count," Ruggiero says. "The key to trading Elliott lies in not getting hung up on the correct wave count, but rather in determining the count that has the least penalty for being wrong."
Finding the correct count requires time. There are nine different wave patterns or degrees of trend in Elliott Wave ranging from the grand supercycle lasting hundreds of years to the sub-minuet degree covering a few hours. Elliott practitioners can spend days arguing over correct wave count but, in many cases, the number will not be confirmed until after the fact.
However, the trader is not interested in whether the chosen index or future is the first or third wave but rather what his or her risk of being wrong is versus the potential reward. A trader is really looking for an entry price that is close to support, which, if broken, will nullify the pattern and result in a small loss but, if correct, will return three to five times the amount risked.
For example, if, in the complete Elliott Wave below, the trader mistook the bottom of wave 2 to be the bottom of wave 4 and entered a long trade, he or she would catch wave 3 instead of wave 5 and still make a good profit because both waves 3 and 5 are generally powerful up moves. In certain cases, a wave 3 is the longest wave in the pattern.
Now let's say the same trader mistook wave B for wave 1, and then entered a long trade at the next pause because he or she thought it was a new wave 3; this pause would've actually been a continuation of wave C, making the trade a painful experience, especially if wave C was incomplete.
In Figure 1 below, we see an example of a wave pattern that was identified by the computer as an ABC wave but was actually part of a much larger corrective wave. It worked out well for the trader, who, instead of earning the expected profit of two- to three-times risk (5.5 points), made more than six times that amount.
The point is that it doesn't really matter if the trader gets the wave count wrong. As long as he or she determines the primary direction of the trend, properly differentiates between the primary and corrective waves and uses tight stops and realistic profit targets, trades can still be very profitable.
|Figure 1 – Five-minute chart of the Dow Industrial Average showing a profitable trade and the Elliott Wave Oscillator in the lower window.|
Elliott Wave Oscillator
What can an Elliott Wave computer trader use to gain greater insight into where he or she is in a wave? Create an Elliott Wave oscillator (EWO), according to Perry Kaufman. The EWO is simply the difference between a five-period and 35-period simple moving average, which in Figure 1 is shown as red and blue moving average lines.
In Metastock, for example, the formula for the EWO is simple. To get the display shown in Figure 1, plot the formula below as a histogram:
EWO = Mov(Close, 5) – Mov(Close, 35).
Note the magenta lines in the main chart and those in the lower EWO window slanting in the opposite direction. This shows clear divergence between price and the Elliott Wave oscillator - a sign that a change in direction is imminent. Kaufman says that a new upward trend is identified when the EWO makes a higher high than the previous EWO high. For example, in an uptrend, a wave-3 EWO high would be greater than a wave-1 high.
As we see in Figure 1, the EWO, like any good oscillator, can also be used as a warning of divergence and the change in direction. After watching the EWO for a while, you will begin to see the pattern. In an uptrend, the EWO will put in a series of higher highs after which it will drop below zero, which will be the ABC corrective pattern. A new series is then about to begin.
Trades confirmed by an oscillator are lower risk than those without confirmation. When the oscillator begins to put in a series of lower highs while price puts in higher highs, get ready for a trend change.
Rather than try to "train" the computer to perform the complex and subjective task of accurately identifying all aspects of the Elliott Wave, it is far more feasible to isolate patterns that are close to each other and places where the penalty for being wrong is minimized.
This means identifying the primary trend, taking trades in this direction and setting tight stops in case you have made an error in your analysis. It won't matter that much if you mistakenly identify one part of the wave for another as long as they are similar parts in the wave cycle.
To help confirm the proper entry and exit points, the Elliott Wave oscillator can be used to choose higher highs and higher lows in an uptrend, or lower highs and lower lows in a downtrend. Divergence between the oscillator and price is also a very useful tool for trade confirmation. Furthermore, wherever possible, confirmation in different time periods - for example, a five- and 15-minute chart for short-term traders, or a daily and weekly chart for longer-term traders - further increases the chances of a profitable outcome.
With a basic understanding of the theory and a bit of practice, it won't be long before you are using what you have learned to enhance your trading acumen.
Next: Elliott Wave: Solving The Probability Problem »
Table of Contents
comments powered by Disqus