In designing a trading model we have to make certain choices that ultimately will influence its performance behavior.
From our perspective, we consider two key properties in designing a good trading model, the resilience of returns over time and their volatility. We believe that an unbiased systematic trend-following strategy can provide a conservative approach for trading volatile futures markets. By their nature, such systems have shown good resilience over time (They have a recurring tendency to recover back from drawdowns). However, such models which have been optimized for offering resilience may not offer a smooth ride. Efforts in designing trading models should aim to produce as consistent returns as possible. This may, however, come at a cost of not acting as good a hedge in times of crisis.
Directional trading models:
Directional and systematic models provide a good strategy for medium to long-term investment horizons. These strategies can be good in generating large profitable trades since they generally hold a trade for a long period. However, they may also produce large drawdowns that can last for long periods if there are no strong trends in the market.
To improve the volatility of returns of such directional, trend following systems we consider the following adaptations:
- Overlays and filters.
- Profit taking whenever directionality and price volatility reach extremes.
- Adjusting of overall portfolio exposure (margin usage versus cash). Typically, “pure” trend programs are not allowed to go to cash and always commit a predetermined risk per a given market.
