Monday, July 1, 2013

Mean Reversion revisited: S&P 500 (What changed in 1987?)

18 Months ago I posted an article about my mean reversion trading in the S&P 500 and a larger view on the history with index data. The data shown is a composite of 8 different strategies in the S&P 500 I monitor. Some of them I trade with the eMini S&P Future. The data used here is daily index data from Yahoo.

Most striking was that the overall systematic approach to mean reversion trading was not working before '87. Especially between 1963 and 1973 it was horrible - a system that you would like to trade, but exactly the other side of each trade. Something that I still cannot explain. But obviously the structure of market movements changed strongly. Thus mean reversion in equity trading is something everyone trading should monitor closely.  

The updated long term chart:
S&P 500 Index - Mean reversion composite





This looks still quite compelling. The big picture is still on track. Zooming in to the recent out-of-sample period......
S&P 500 Index - Mean reversion composite

Still an uptrend but the it looks a bit choppy now. This is also what I experienced by trading mean reversion systems in the last 2 years. Mean reversion needs a larger range for worst case stops, otherwise the expected return suffers. Thus in case of a stopped out trade it is quite hurting the performance as in January 2012 and April/May 2013.

Looking at the distributions....
S&P 500 Index - Mean reversion composite / Distribution of returns

.... and the numbers*:














What is the conclusion out of this? As a lot others in the quant trading blogosphere noted before: The best times for mean reversion systems seem to be over. Pure, unfiltered signals are even more problematic since 2010. I slightly reduced the exposure to mean reversion in my trading portfolio in the last year. But I am still trading 'my' markets monitoring the results and risks closely.   


* please note that the numbers differ a bit from the former post. In addition the Max DD % 'After Crash '87' was wrong/too high in the last post.


Disclaimer

The information on this site is provided for statistical and informational purposes only. Nothing herein should be interpreted or regarded as personalized investment advice or to state or imply that past results are an indication of future performance. The author of this website is not a licensed financial advisor and will not accept liability for any loss or damage, including without limitation to, any loss of profit, which may arise directly or indirectly from use of or reliance on the content of this website(s).  


Under no circumstances does this information represent an advice or recommendation to buy, sell or hold any security.

5 comments:

Anonymous said...

What's up, yup this piece of writing is actually
good and I have learned lot of things from it about blogging.

thanks.

My blog - moving company Massachusetts

Stefan Martinek said...

What is the cost of trading?

Andreas Marquardt said...

Stefan, I backtest the S&P 500 like trading the eMini. Means 0.25 PTS per side, 0.5 roundturn.

Jared Broad said...

Great post, what back-testing platform do you use?

We're building a free C# one with free intraday data back to 1998 https://www.quantconnect.com - I think you'll like it :)

Andreas Marquardt said...

Jared, I use Multicharts mostly for futures (like here) and Amibroker for stocks and rotational trading.