Jaws Mean Reversion Strategy

In this post I’ll review a simple strategy that I developed many years ago that attempts to exploit reversion to the mean. I’ve named the strategy Jaws due to the Jaw like pattern that the indicators exhibit before an entry condition occurs.

The rules:

Buy on the close when the 2 period moving average closes 3% or more below the 5 period moving average.
Sell on the close when the 2 period moving average closes above the 5 period moving average.

That’s it! When it comes to building trading systems, simple is always better. Complicated strategies are susceptible to over-fitting past data and prone to failure in the future. Too many rules and conditions deteriorate the strategies ability to adapt to new market conditions, decreasing the likelihood of success in the future. Below I’ve included an illustration of what the typical trade looks like, it should become apparent where the strategy gets its name – when the 2 period moving average begins to move away from the 5 period moving average it’s analogous to a jaw being opened. Once the gap between the two moving averages exceeds 3% an entry condition occurs, and when the 2 period moving average moves back above the 5 period moving average an exit condition occurs.

This simple strategy does a remarkable job of capturing reversion trades. Since 2000 it’s completed 3040 trades, has a winning rate of 69.52% and has generated positive returns every single year since 2000. (The strategy was applied to the Top 60 JSE listed equities ranked daily by liquidity from 2000 to the present). Below is an illustration of the strategy’s equity curve.

It’s possible to improve upon the strategy by adjusting the parameter sets used in testing. For instance, one could demand a larger gap between the two moving averages before entering a trade, such as 5% as opposed to 3%. By enforcing more demanding entry conditions one can improve the single trade performance, but at the cost of reducing trade activity (for every parameter change there’s a trade-off that has to be made). Active traders could employ less demanding conditions to boost trade activity. Further, additional simple overlays could also be applied to improve performance, such as a trend filter. The variations are unlimited, the key is to build and trade a strategy that fits your investment objectives and risk/return profile.

Our QuantLab boasts thousands of strategy variations that traders can access. QuantLab offers you the ability to view countless historical performance statistics related to each strategy, build and backtest your very own portfolio and execute the daily signals related to your portfolio with QuantLab’s integrated execution capabilities.

Happy Trading,

Passionate about generating and sharing quantified trading models that empower individuals to trade successfully. I founded www.sutherlandresearch.com to realise my passion.

Leave a reply:

Site Footer