BACK TESTING

[ianstevenson]

Hi all,
2 questions re back testing a trend following system, aiming to catch the likes of A2M and IEL.

In the “setting / portfolio” section there’s an option “limit trade size as % of entry bar volume”.
I think I’ve got that covered by coding my position size as max 1/20th (5%) of average daily volume. So should I set this as 0 or 5. The results vary greatly with different settings.

What time periods do you use when back testing? Do you look at 10 years history 1 or 2 years at a time, 5 years, 20 years?

Cheers Ian

[Nick Radge]

Ian,
Regarding your first question, I’m not sure if you’re doubling up or not. I don’t use it in the code, but I do use it in the backtester. Perhaps Craig can add some insight.

Backtest longer periods. 10-years is an absolute minimum. I like going back to 1995.

Nick

[ianstevenson]

thanks Nick, and how long are the periods you look at in that history , 1 yr, 3 yrs 5 yrs?. I guess i would have discarded something if it didn’t work for 3 yrs.
cheers Ian

[Nick Radge]

Well, you need to understand what and why the system is doing what its doing. A long only trend system in equities won’t produce returns during a sustained bear market, but that doesn’t mean it’s broken.

[Author]

Posts