Skip to content

Testing Multiple Strategies

July 3, 2009

Off late, I had been experimenting quite a bit with how multiple systems interact with each other, on a portfolio level to deliver returns.

The reason, I am dealing with this, is because incorporating a strategy of just one flavour can severly hamper the equity curve’s quality. The right way is in mixing systems of various flavours and possibly some special strategies on specific instruments so that equity curve has a more smooth output.

If I want to test a trend following system on all the stocks of SPX but a counter-trend system on the index itself, together,common sense tells us, that the resultant equity curve will be the linear combination of seperate equity curve of the two strategies. But this is flawed reasoning. There is something called Synergetic Effect and Antagonistic Effect in nature. As the name says, synergetic effect means, the resultant being better than the sum of individual components. And by analogy the antagonistic effect. As a trader/fund manager, its imperative that a system hedge is performed. Meaning, a system is hedged by another system thus making the equity curve better, not only in terms of drawdowns, but also in terms of returns.

Consider I have 100 bucks in my account and I have allocated 60%-40% ratio to a trend following system and a counter trend following system. Assuming their drawdowns are not in sync, then we can have a compelling case of higher returns. How ?It effectively has an effect of compounding faster than any single strategy, with lesser drawdown[only if the equity curve have a phase difference, a lead or a lag]. Thus increasing the CAGR/Max DD ratio.

So in this quest, I went to Amibroker, one of the softwares I avidly use.I use it mostly for coding my systems, but what irks me, is its absolute apathy towards multi-strategy testing.You cannot run a combination of two or more strategies together. And what surprises me is the lack of such feature even in Wealth Lab. I mostly am at a loss of overcoming this drawback. Possibly, one idea I can get is, catch hold of the two equity curves, and scale the curves for the beginning of each month, at the end of the month, the returns they will be showing, pool it back, and again scale the curves, show the returns. And the effective summation of their scaled returns of the equity curve might give it a good chance of simulating the same in real life.

Sigh! 🙂


No comments yet

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: