Technical Trading Systems
Unfortunately, it did not display any potential of being profitable during the in‐
sample test period. In this example, none of the moving averages associated with
equity curves A, B, or C would represent wise choices for a moving average cross-
over system.
A trading system said to be robust if it has the ability to maintain consistent
positive expectancy:■ (^) Under all market conditions
■ (^) In a wide range of markets
■ (^) With variations in its parameter settings
Although it is universally agreed that a trading system should be robust under
all market conditions, there is some debate as to whether a system has to also be
robust in all markets, as there is nothing wrong or disadvantageous about designing
a system to trade a specific market. Another common test of robustness is the ability
of the system to perform consistently well with changes in its parameter settings. For
example, if small adjustments in the lookback period of a moving average crossover
system produce large volatility in the equity, we say that the system is less robust.
29.4 Performance Measurement
It is important to be able to gauge the level of performance of a trading sys-
tem in order to be able to analyze it and effect improvements. Below are a few
performance‐ and trade‐related metrics that help traders keep track of how the
system is behaving:Figure 29.4 Equity Curve Displaying Extreme Drawdown During the In‐Sample Test
Period.