WFM-which values are the right ones?
11 replies
Patrick
11 years ago #114235
Hi my friends,
please help me. I read story about WFM already.
Check the attached picture please and tell me, how do i know which values are right ones? If i change them, the result can change from passed to failed. Which values should i use? these from article? Or some different?
Thanks for advice,
Patrick
0
Tomas Matejka
11 years ago #132833
I see nothing wrong with settings you provided. This can serve well. If you have more computing power or are patient you can go for more WF runs too
0
munchie
11 years ago #132839
Good question. I feel there should be some support as to what a ‘typical’ working range or value should be, no matter how wide or narrow. From my experience, it’s one of the biggest confusions and this applies to things like stagnation, drawdown, return/drawdown etc.
Sent from my iPhone using Tapatalk
0
Patrick
11 years ago #132898
if i use setting from the article the result looks like this one-attached. so it means the strategy, when it passed MC analyse, is robust, right? (chart is not 3D i dont know why…)
0
Patrick
11 years ago #132923
I see nothing wrong with settings you provided. This can serve well. If you have more computing power or are patient you can go for more WF runs too
I have problem, all strategies i have pass the simulated WFM test. Is the setting weak/not hard?
Question to others, if you use setting from the article, pass your strategies or not?
Thank for help, Patrick
0
munchie
11 years ago #132924
How many strategies did you test?
Sent from my iPhone using Tapatalk
0
Patrick
11 years ago #132935
How many strategies did you test?
Sent from my iPhone using Tapatalk
min 15 strategies, all passed WFM with setting from article…
0
munchie
11 years ago #132936
I would adjust the settings incrementally and retest and continue doing this until some start to fail….once you’re left with 1 or 2, you will know the settings are about right and the strategies you’re left with are the best
Sent from my iPhone using Tapatalk
0
Patrick
11 years ago #132937
I would adjust the settings incrementally and retest and continue doing this until some start to fail….once you’re left with 1 or 2, you will know the settings are about right and the strategies you’re left with are the best
Sent from my iPhone using Tapatalk
But how can i make the conditions harder? Do you have any other setting or experience? Or someone other?
Thanks
0
munchie
11 years ago #132950
But how can i make the conditions harder? Do you have any other setting or experience? Or someone other?
Thanks
My experience is only limited so maybe more experienced guys can provide some useful advice. My experience when following the WFM was the same as you. You can consider changing the number of walk forward runs. It’s a case of trial and error IMO, and I think my filtering process is best to further define the single best strategy and settings for WFM. When everything fails or everything passes then you should be able to narrow down the range. I’ll have a look when I’m back at my system later today.
Sent from my iPhone using Tapatalk
0
munchie
11 years ago #132962
Patrick, you can vary the settings in WFM and as every strategic is different, look at the source code to find out what indicators it uses, and maybe adjust parameters for those and the stop loss. Don’t make extreme changes to everything, small changes in things that are important to the strategy is enough. Remember, with WFM you are trying to find out if the strategy benefits from periodic reoptimisation.
Sent from my iPhone using Tapatalk
0
Patrick
11 years ago #132974
i tried different set also, but result is still good at first picture.
when i try to optimize only SL, PT and 1 indicator, result is the same (second pic)
i do WFM test to prevent over-optimizing.
0
Viewing 11 replies - 1 through 11 (of 11 total)