E-mail:    

 

TRADING SYSTEMS

Simple Mechanical Trend Following in the Forex Market

Simple, Mechanical, Trend, Following, Forex Market, Trading Systems, fx trader, forex

Trying to profit from sustained momentum is one of the most commonly used and all-time successful strategies in trading. However most traders have problems implementing such systems because they over-complicate the notion of trend-following, often using very convoluted approaches that focus on very complex approximations that generally increase the data-mining and curve-fitting biases present within the strategy creation process. In this article we are going to evaluate a very simple trend-following approach across six different Forex trading instruments simulated over more than 28 years of market data.

If you want to create the simplest possible trend-following approach you need to use the simplest measure of directionality available – open-to-open price differences – and avoid the use of any additional exit parameters besides a simple stoploss used for risk control. We can imagine a very simple system that follows some very basic rules to follow trends on the daily timeframe:

 Long Signal:

  · Open price today is larger than open price N days ago

 Short Signal:

  · Open price today is lower than open price N days ago

 Rules:

  · If no trades are open and there is a long signal then enter a long trade.

  · If no trades are open and there is a short signal then enter a short trade.

  · When a new trade is entered place a stoploss SL% of the ATR-20 away from the current entry price.

  · Lot sizes of new open trades are calculated such that the loss at a touch of the stoploss is 1%.

  · If there is an open trade in one direction and we receive a signal in the same direction then move the stoploss as if a new trade had just been opened at the current open price.

  · If there is an open trade and we receive a signal in the opposite direction then close the current trade and open a new trade in the signal's direction.

The above system has several desirable characteristics. It contains only two parameters, the lookback period (N) used to measure the trend's direction and the stoploss size (SL) as a percentage of the daily ATR-20 indicator. Notice that the system can only have one trade open at a time and there are no exit criteria besides the stoploss and the reverse signal. Another very important characteristic of this strategy is the updating of the stoploss when signals in the same direction are generated. This in fact makes our stoploss act like a trailing stop without the need to define any additional parameters. The very small number of parameters and their multiple usage across the strategy ensures that the strategy remains robust to both data-mining and curve-fitting biases since only very limited modifications are possible due to the strong restrictions in degrees of freedom.

Next>>