|
PureBytes Links
Trading Reference Links
|
I think you answered your own question for the optimizing you did.
You did exactly what you are supposed to do. You built and tested a
system on one set of data and then went and tested on an out of sample
set of data to see if it was truly a good idea or simply
"fitted". You found your answer.
I use optimization selectively. I don't try to optimize every single
variable in my system right from the start. I try to develop concepts
that seem like they would makes sense. From there, I go on to test
other data. If that holds some promise, I go back to the original
data and may try to optimize a few of the variables. Once that is
done, I go back to the other data and see what happened. Finally,
when I think I've got something, still new data is introduced. Does
it still work?
It is important to select the right data too. Comparing a system on
say a volatile tech stock to that of some tight ranged utility stock
makes little sense. These have different personalities.
So, yes, optimizing does have its place but you have to know what you
are doing and what the dangers are. This is not an easy task. It
takes time and effort.
Pring
>
> Is optimizing bullshit?
>
> I have made systems which perform quite well, but when i try them
on
> multiple stocks the results sometimes change from positive to
> negative.
>
> If i optimize i can get a positive result on this security, but my
> question is do these optimizing values hold for future signals, I
> think they dont as they simply fit the test data not the future
data
> which offers endless possibilities.
>
> so is it useless to optimize?
>
> toni
>
>
> Yahoo! Groups Sponsor
> ADVERTISEMENT
>
>
>
>
> To unsubscribe from this group, send an email to:
> Metastockusers-unsubscribe@xxxxxxxxxxx
>
>
>
> Your use of Yahoo! Groups is subject to the Yahoo! Terms of
Service.
To unsubscribe from this group, send an email to:
Metastockusers-unsubscribe@xxxxxxxxxxx
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|