Page 1 of 1

Performance over historical time frames

Posted: Sun Feb 01, 2004 11:49 pm
by Bruce
I have been doing considerable historical testing with the various systems. What is interesting is the performance gets chopped away with every new time frame.

For instance for one of the systems, the periods of:

1970-1980 =CAGR OF 70%

1980-1990= CAGR OF 58%

1990-2003= CAGR OF 20%

i am certain this is no revelation for any of the testers here, but I did find it interesting. Also interesting is the drawdowns go up with every time frame.

Anyone have any reason why? What has influenced the markets to weaken the performance over these time frames? Program Trading?

Posted: Mon Feb 02, 2004 7:48 am
by Jwebster
While i don't know what type of system you are using I would attribute a lot of the difference to the massive increase in volatility. If you don't have system that maxmises this or screens it out then I imagine you are going to get chopped up more frequently as your timeframe approaches the present day.

Posted: Mon Feb 02, 2004 2:27 pm
by edward kim
Bruce,

You have provided only your CAGR, which is not enough information for us to give you comments. How about MAR, Sharpe, drawdown levels, the markets you trade, the type of system you trade, etc. ?

My trading system consistently performs from the 70s all the way through to today (in all return and performance ratio categories). So I have an idea of what at least two things might account for your results, but I won't make any guesses until I see more information.

Edward