Charting the Stock

Market For Profit
Building client wealth SINCE 1998

 The Chartist trial green

  • Parent Category: Articles

Combining Strategies For Comfort

Combining Strategies For ComfortBefore I get into today's article I need to respond to some comments received after last week's Mean Reversion article. One criticism was the system was losing its edge as the equity curve had been flattening over the last few years.

This most likely came about as I used a log scale chart. So for better clarity, here is the linear scale chart.

fixed percentage allocation

One of the big demons newer or less experienced traders face is lack of equity growth even if their strategy has a proven long term edge.

Lack of growth can be from drawdowns, or where the equity curve flat lines for extended periods of time.

Both are natural nuances of trading and investing, regardless of the strategy used. Whilst some accept the 'terms and conditions' and take the long term view, others are more susceptible to the Beginners Cycle.

We've all heard the benefits of having a diversified portfolio of shares.

I, however, prefer to have a diversified portfolio of strategies. Doing so 'tends' to smooth equity growth, lower drawdowns and help alleviate flat periods.

In summary, having a variety of systems can greatly reduce trader frustration.

The chart below shows a 50/50 combination of the mean reversion strategy and a trend following strategy, specifically the Growth Portfolio which Trish and I use to manage our retirement account.

buy and hold combined

On first glance the combined performance is not as good as the mean reversion strategy on its own (see first chart). The CAGR falls from 19.1% to 17.2%.

However, the drawdown also falls from -15.3% to -10.5%.

The correlation of the two strategies is 0.0915 - almost completely uncorrelated.

To put this into perspective, the trend strategy makes money in 29 months where the mean reversion strategy loses money. And the mean reversion strategy makes money in 69 months where the trend strategy loses money.

In other words, of the 251 months of testing shown above, almost 100 months are uncorrelated.

And it's for this reason I trade 5 strategies capturing different styles, different time frames, different markets and even down to different universes within a market.

Try before You buy The Chartist 200