A strategy's life-span...

Hi!

Couple of weeks ago I somewhere read (can't really remember where I read or who said it) that strategies don't have infinite life - a strategy maybe works for 4-6 years or so.

A thought occurred to me consindering this. Often when I backtest my srategy the backtest window often consists of 10-12 years or so. Does this mean that the strategy "doesn't really work" on half (or so) of the backtest window? Won't the result be missleading then?

And, isn't the optimization-function supposed to take care of this "problem" since then optimization helps the strategy to adapt to new market conditions?


MS

What timeframe do you use for trading? Were 4-6 years meant for a strategy with parameters which are never re-tuned? Or self-tuned during strategy's life?

Even using 10-12 years for backtest, you could see when equity curve went up and when down. So if it went up only once 8 years ago and then went sideways - that's a bad sign i.m.h.o. Chances are high, that there was a big change in those years and, yes, the strategy "doesn't really work" in a considerable part of backtest window. That's why I beleive that the window whould'n be too big for optimization; and that OOS test is a must.

Time span choosen for optimisation matters a lot. E.Chan:

Hi!

Mostly 60 or 240 min.

I'm mostly talking about self-tuned strategies.

With some systems, the reason why it expires is easy to analyze, with other systems it's hard to tell. But yes, most systems indeed expire after about 5 to 6 years, at least in our experience, even when the system is regularly optimized. That's why it normally does not make sense to run a backtest over more that 6-7 years.