Reply

Giving up on workflow

3 replies

dsilva743

Customer, bbp_participant, sq-ultimate, 12 replies.

Visit profile

3 years ago #279244

After creating a workflow and letting it run for a few days or weeks, would you just start from scratch again or just switch things around?

0

tightpips

Customer, bbp_participant, community, sq-ultimate, 14 replies.

Visit profile

3 years ago #279256

It’s always tempting to use the standard build because you see results faster, but for a more robust process, workflow is the way to go.

you may want to adjust your filter a bit if you not getting much results, I use random gen to get a batch of strategies then put them as the initial population for 2nd genetic generation workflow. it consists of 2 workflows with higher qualifier filters and additional building blocks at the genetic workflow.

 

0

dsilva743

Customer, bbp_participant, sq-ultimate, 12 replies.

Visit profile

3 years ago #279258

Correct me if I’m wrong, so when the strategies are created from random gen, you add a new task, take the strategies that were randomly generated, and improve the existing strategy with genetic options?

 

that is an interesting way of going about it. For some reason, my filters are not that strict.

-> IS 10years profit factor > 1.3

-> IS 10years # of trades > 500

-> IS 10years winng % > 30%

 

-> OOS 5years profit factor > 1.3

-> OOS 5years # of trades > 250

-> OOS 5years winning % > 30%

 

Im not sure if my issue is selecting the correct building blocks for lets say trend following, or its my filtering in rankings. i don’t use filters in genetic options

0

tightpips

Customer, bbp_participant, community, sq-ultimate, 14 replies.

Visit profile

3 years ago #279280

Yeah, I use random to just get a glimpse of how the building blocks I selected are going to do. Once you have, lets say 200-300 good strategies, I’ll put them into the initial bank for genetics.

 

That really depends on how you select your building blocks, I would start with turning off the “Auto delete” function on the filter, then manually check the main reasons why your strategy fails. Aslo the progress status page that shows you the overall reject number will give you a good indicator of what’s failing the most.

 

As for filters, it really depends on what your preference is, everyone got their own style of trading and tolerance of risk. I tend to focus on % stagnation and Mas DD %, not so much on win/lose ratio.

 

Lastly, do not look for the silver bullet strategy, you will need to get a portfolio going on in order to be consistent. Single strategy will push and test your emotion to the max.

 

Hope this helps.

0

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