I did backtest my formula

gkishot said:
I disagree. There is no way around backtesting your strategy before going live.
More...
With all due respect, Yes and No.
Backtesting is only one of the first steps. There are many more following that (before going live).
And backtesting results by itself is meaningless if you don't do the next validation steps (as the OP did).
 
rb7 said:
With all due respect, Yes and No.
Backtesting is only one of the first steps. There are many more following that (before going live).
And backtesting results by itself is meaningless if you don't do the next validation steps (as the OP did).
More...
You sound stressed, bro.

V8WGpct.gif


Once again, semantics.

All those 'other validation steps' are included under the term: 'backtesting.' Then there is forwardtesting.

Backtesting is the prelude to forwardtesting. Backtesting is not meaningless.

Even if an individual decides not to backtest, that doesn't imply the general statement "backtesting is meaningless." Even if it is meaningless to one or more individuals.

Are drug, clinical trials meaningless? Of course not. Just because there are more steps, doesn't mean the first steps are meaningless.

If scientists are happy about their clinical trials, does that mean message board communities get to assume that they are done testing? Of course not.

Good luck OP!
 
Seems like some ET participants have difficulties reading and/or understanding basic English text.
I've never wrote "backtesting is meaningless".
Whatever bro.
I'll exchange your joint for a cigar.
No offence.
Long live to the OP!
 
rb7 said:
Seems like some ET participants have difficulties reading and/or understanding basic English text.
I've never wrote "backtesting is meaningless".
Whatever bro.
I'll exchange your joint for a cigar.
No offence.
Long live to the OP!
More...
To be fair, I'll just drop your quote right here:
rb7 said:
Anyone who think backtesting results has any meaning is a fool, or a newbie, or both.
More...
Cigars are good.

giphy.gif
 
janbillian said:
yinyang, backtesting can be valuable if you don't fall into the trap over curvefitting, proabably the #1 error. A dataset for a give timeperiod gets an indicator and / or candelstick overlay, and than depending on adjustment possibilites, the setups caught the past timeperiod perfectly.

#2 error is being too conservative on slippage assumptions. Maybe the fills would have been worse.

To show the value of your system, you should try an out of sample test with your system and the given parameters. How would it have performed between say 2006 - 2013 ?

On paper, it looks perfect, but you'll be hard pressed to find traders who can generate 30% per year or more in real life.
More...

Thanks for the comments. I used 4h candle which only available up to Yr 2013. Any advise where I can get S&P futures 4h price data?
 
yinyang said:
Thanks for the comments. I used 4h candle which only available up to Yr 2013. Any advise where I can get S&P futures 4h price data?
More...

Yinyang, I have one URL for you, https://firstratedata.com/i/futures/ES . You can you use any online search engine, many resources are available for long term data, but they of course charge you for it. Free data is unfortunately very limited. If your system performed that well with the exact settings in the six years before, you may have a winner, however there's no way you'll realize a 60% cagr p.a. . But 20 - 25% p.a. may be in the cards.
 
You must log in or register to reply here.
Back
Top