Backtest Time Frame - newbie question

Quote from NoWorries:

I think you also might want to ask yourself the question after how many days (some number between one and never) of trading you're planning to perform another set of backtests, and perhaps calibrate.

Then you might want to do a series of walk-forward-tests using all your data (e.g. 4000 bars) and different values for your optimization period X (e.g. 400 bars), and your (hypothetical) trading period Y (e.g. 40 bars), see if any patterns emerge and look for more-or-less robust combinations of X and Y, judged by a variety of statistics of interest (e.g. annualized return, dd, sharpe etc.), estimated over the (hypothetical) trading periods (not over the optimization periods!).

This still doesn't guarantee anything for the future, but it's probably more informative than your current approach. E.g. if your system shows acceptable outcomes for a broad range of X and Y values, you might have something valuable. If it's very sensitive to X and Y, you might want to try something else.
More...

I will give it a try this afternoon

Thanks
 
Quote from OddTrader:

You must be saying they are too good to be true. :D
More...

There are two results I am looking into:

1.) 84% at 250 bars (Top 6 in profit), but 53.5% at 4000 bars (Top 7 in profit)

2.) 74% at 250 bars (Top 23), 59.82% at 4000 bars. (Top 6)

There are 1960 tests for this system.

I am confuse!
 
Quote from Willleung:

There are two results I am looking into:

1.) 84% at 250 bars (Top 6 in profit), but 53.5% at 4000 bars (Top 7 in profit)

2.) 74% at 250 bars (Top 23), 59.82% at 4000 bars. (Top 6)

There are 1960 tests for this system.

I am confuse!
More...

Looking at % return alone would be the very first step advancing to further hard works. :D

Search and study. Good luck to your myth finding journey! :)
 
Quote from OddTrader:

Looking at % return alone would be the very first step advancing to further hard works. :D

Search and study. Good luck to your myth finding journey! :)
More...

Yeah, will lock myself in the apartment for a few more days (months?)

LOL...


Doing the Walk Forward Test now, what should I be looking for between the Optimization Period and the Hypothetical Period?
 
Quote from Willleung:

Yeah, will lock myself in the apartment for a few more days (months?)

LOL...


Doing the Walk Forward Test now, what should I be looking for between the Optimization Period and the Hypothetical Period?

ie, Top X & Y in the optimization periods hold up in the hypothetical period?
More...

Only Hypo can answer Hypo questions! :D
 
Permit me to share a dirty little secret. If a set of rules don't test with positive expectancy after commission and slippage with NO optimization, there is nothing there.
 
Quote from Willleung:

Yeah, will lock myself in the apartment for a few more days (months?)

LOL...


Doing the Walk Forward Test now, what should I be looking for between the Optimization Period and the Hypothetical Period?
More...

Years! Use optimization to make sure your system parameters are in well behaved regions.
 
Quote from OddTrader:

Jack's telling people that his methods cannot be backtesed and backtesting is not useful would be a good way to drive people to pay more attention and follow him without too many questions.

Perhaps he uses backtesting at home frequently. :D
More...

Jack can still be with us in spirit :D Here are my two favorite Hershey quotes on backtesting:

Backtesting is not a viable means for finding anything out. The reason why is fairly straightforward to determine. If you are backtesting something and you feel you have grasp the thing you are backtesting, then you have failed to really grasp what (the thing) is going on.

and

It is not possible for a person who does not know what is going on to back test this stuff. I have stated it. And it just flows from there using boolean algebra.
 
You must log in or register to reply here.
Back
Top