Tuesday, July 15, 2014

Back Testing Moving Average Trading (2)

Just to review - the ultimate objective is to determine one of three conditions – is it going up? Is it going down? (and) When will it stop doing that (when will it turn)? Some technical studies will lead – like momentum, relative performance and divergence. Some are coincident – like trend lines and some lag – like moving averages. Simple moving averages are popular but they are lagging trend-following studies and should be used to confirm a trend and not to make a trading decision. This time we look at multiple SMA crossovers.

I did more back-testing on the S&P500 using a two simple moving average (SMA) cross-over – in this case a 5 and a 15 period with the difference smooth by 3 and displayed as a histogram. The trading was on the long side only – sell on a negative histogram and buy back on a positive histogram. The objective was the beat a buy-and-hold strategy over the same period.

The 5/15/3 SMA cross-over monthly from 3/29/1991 to 7/11/2014 - buy and hold got us +424% and SMA cross over trading got us +537% over the same period. We had 6 profitable trades out of 6 signals for a perfect trading record.

Conclusion – we know that signals based on price and SMA crossovers generate multiple false signals – but signals based on two or more SMA crossovers generate reliable signals – in this example having us avoid the 2000-2002 and 2008 bears using long term monthly data. Our long term S&P500 chart displayed here clearly plots the 5/15/3 histogram with the buy & sell zones.

No comments: