Reply

WFM – Fixed size, or fixed $ amount

3 replies

nolube

Customer, bbp_participant, community, 115 replies.

Visit profile

8 years ago #113522

Hey SQers, I ran two WFM on the SAME strategy.. One had a fixed lot size of 1, the 2nd one had a fixed $ amount risk of $200. They both got wayyyy different results. On default settings, the ‘fixed $ risk’ passed, whilst the ‘fixed lot’ test failed. Can someone please help me get my head around this?

 

Both strategies used exactly the same settings and data!

 

 

Data is GBPUSD, H1 timeframe. Strategy attached also.

0

nolube

Customer, bbp_participant, community, 115 replies.

Visit profile

8 years ago #129513

So I’m just thinking aloud but fixed $ obviously must be more robust as it survives different market conditions (volatility) better.

 

So therefor, is it better to test WFM on fixed lot as a further robustness test?

0

seaton

Customer, bbp_participant, community, 161 replies.

Visit profile

8 years ago #129535

Ive been thinking about this last night and this is what the conclusion Ive come to.

 

With fixed $ your risk is fixed across all different tests and strategies regardless of what the SL mechanism is, so really the metrics produced aka the results such as profit factor etc will be more comparable across all tests and different strategies.

 

Where as fixed lot testing will have different risk depending on their SL mechanism so in reality the metrics produced aren’t really comparable, and will change if you optimise the SL

 

In the case above the strategy being tested may be fine with the risk being under test on the fixed $ but is unstable on a fixed lot size because the risk becomes unacceptable, i.e the losses are larger losses.

 

My thoughts are that for mining strategies fixed lot is best because you want to see what the results are mined in a general sense, and are then filtered down to your candidate strategies, then for improving, WFM and optimisation fixed $ because you need to compare the results from one run to another on fixed risk, also if you do this for all your strategy testing over the long run you have a standard way of comparing them.

 

I may be completely off target here so am happy to open up for discussion to see what others think.

 

Stephen…

0

Matusiak Adrian

Customer, bbp_participant, community, 300 replies.

Visit profile

8 years ago #129570

I won’t tell anything about methods, but I would like to pay Your attention to chart of WFM of strategy that passed. 

 

nolube, please check also “WF Drawdown Stability” – surley Your strategy won’t pass the WFM. 

btn_viewmy_160x33.png

0

Viewing 3 replies - 1 through 3 (of 3 total)