Reply

RT randomise parameters shows only one equity

7 replies

Gui

Customer, bbp_participant, community, 59 replies.

Visit profile

7 years ago #116525

Hello gentlemen,

 

I would like your opinion on the randomise parameter test using 20% and 20% as parameters for 200 runs.

I am testing some Daily strategies, with that function, and 95% of the tests comes with only one equity line instead of 200.

 

‘‹My question is that, from an SQ point of view, do we have 200 times the same equity line (plausible) or the 199 tests having no equity line are then not pondered in the 95% confidence (obviously)?

What would be your take then on having a strategy that is not affected AT ALL by the randomise parameter (could be an open at new bar when CCI>0 for example)?

 

Cheers!

Think different

0

Karish

Customer, bbp_participant, community, sq-ultimate, 443 replies.

Visit profile

7 years ago #142211

Thats a good question, i was noticing that sometimes when i do MC tests some strategies are not showing any other equity curves then the original one, but it will show the whole table of the 95% etc..,

it is a bug that happens when using too much MC runs,

so when i use lower MC runs next time it will show me all the equity curves that should be shown,

sometimes even when i just run the MC another time with the same amount of MC runs it will fix the bug,

 

dont know whats the deal with that problem..,

 

BTW about the 200 times, you should get 200 equitiy curves of MC + 1 that is your original equity curve,

 

about the CCI>0 and other Simple rules, that is also a thing that i would like to hear about 🙂

0

tomas262

Administrator, sq-ultimate, 2 replies.

Visit profile

7 years ago #142261

Hello,

 

when I use simple rule CCI>0 on daily tf I still get multiple equities for RT test (randomize parameters 20% value with 20% probability). The strategy still has stop-loss and target that can be altered. If you attach your strategy I can check that too

0

Gui

Customer, bbp_participant, community, 59 replies.

Visit profile

7 years ago #142264

Hello,

 

when I use simple rule CCI>0 on daily tf I still get multiple equities for RT test (randomize parameters 20% value with 20% probability). The strategy still has stop-loss and target that can be altered. If you attach your strategy I can check that too

Think different

0

Gui

Customer, bbp_participant, community, 59 replies.

Visit profile

7 years ago #142265

Hello,

 

when I use simple rule CCI>0 on daily tf I still get multiple equities for RT test (randomize parameters 20% value with 20% probability). The strategy still has stop-loss and target that can be altered. If you attach your strategy I can check that too

Thanks Tomas, see post above please. I copied 4 out of 26, 5% were working, I hope these aren’t the one I attached lol.

Think different

0

Gui

Customer, bbp_participant, community, 59 replies.

Visit profile

7 years ago #142301

Thanks Tomas, see post above please. I copied 4 out of 26, 5% were working, I hope these aren’t the one I attached lol.

Going through to my backlog, it did happened quite frequently actually in other generations, of any kind too.

Think different

0

tomas262

Administrator, sq-ultimate, 2 replies.

Visit profile

7 years ago #142311

Hello,

 

we have tested Strategy 0.1318428.str and it also does not produce additional equity curves. There seems to be some kind of bug that happens occasionally. This will be fixed with SQ4

0

Gui

Customer, bbp_participant, community, 59 replies.

Visit profile

7 years ago #142322

Hello,

 

we have tested Strategy 0.1318428.str and it also does not produce additional equity curves. There seems to be some kind of bug that happens occasionally. This will be fixed with SQ4

Hi Tomas,

Thanks, I discarded them then, no risk.

Think different

0

Viewing 7 replies - 1 through 7 (of 7 total)