Ãîëîâíà 

Only the equity curve provides a complete and continuous picture of your system's performance over time. The usual test summary tells you little about how your design tradeoffs alter performance on a daytoday basis. Hence your system development is not complete until you understand the impact of your decisions on the evolution of account equity.
In this chapter we take a detailed look at how to measure the smoothness of the equity curve using the standard error (SE) from linear regression analysisthe larger the SE, the rougher the equity curve. Then we see how the equity curve for the 65sma3cc system changes with different exit strategies at the contract level. You will get a feel for how your design choices translate into equity changes.
Next, we discover how SE changes when you combine two systems trading the same market. A common belief is that trading many different markets gives a smoother equity curve. We explore this belief by combining two markets that have some positive covariance.
We then explore the monthly changes in equity curves, examining the performance of the 65sma3cc system trading the deutsche mark over monthly intervals of different lengths. These quantities are termed the interval equity changes. Our goal here is to see how a system does
180 Equity Curve Analysis
over all 1month, 3month or 6month intervals in the test period. These measures help in understanding the effects of adding a trailing stop or changing exit strategies.
These tests show that exit strategies alone do not improve equity curve smoothness (that is, reduce standard error); we look at changing a system's design. Filtering the usual channel breakout system gives a smoother equity curve.
The usual performance summary reveals none of this information, so the new insights from this analysis make it well worth the effort. After reading this chapter you can:
1. Measure the smoothness of an equity curve.
2. Understand the impact of system design on changes in the equity curve.
3. Grasp the effect of diversification on equity curves.
4. Recognize the benefits of using filters in system design.
Summary  Introduction  Channel Breakout on Close with Trailing Stops  Channel Breakout on Close with Volatility Exit  Channel Breakout with 20Tick Barrier  Channel Breakout System with Inside Volatility Barrier  Statistical Significance of Channel Breakout Variations  Day Exit Reference System  Two ADX Variations  The Pullback System 