READ-ONLY

The forum is now a read-only archive.

For bug reports & platform questions → [email protected]

Our community lives on Discord and YouTube — come join us!

Different results on tasks that run retests with the same configurations

0 replies

Sergio La Colla

Subscriber, bbp_participant, customer, community, sq-ultimate, 1 replies.

Visit profile

10 months ago #292369

I get different results on tasks that run retests with the same configurations.
I have a custom project based on my Strategy template. Each task reads from the previous database and saves to a new database. Thask of build, optimization, and retest with cross check always have different results in terms of net profit and number of trades. Typically, the differences are minimal (but I wouldn’t expect them), while on the retest task where I enable cross checking with WF Matrix (no optimation), I get profits up to 4 times higher and 20% more trades. Furthermore, if, after the build task with “1 minute data tick precision,” I run another task on the generated strategy with a “higher back test precision” cross check, also using “1 minute data tick precision,” the test fails because it finds differences of at least 20% in profit and number of trades. What could be the problem? I checked and the configurations are identical. I also used the “Copy Config to task” feature to align all the tasks. I’m new to sqx and this thing is causing me uncertainty that has blocked me. Any ideas?

0