Hi guys I’m opening this thread to discuss about the automatic filter that the new version of StrategyQuant apply when the strategies are generated.
At this moment we have the folowing filter:
-Too many ambiguos trades
-Too many open trades
-No filled trades
-Outlier trade – one exceptionally big trade
-Zero PL trades
-Zero duration trades
-Too little trades
-Too many trades closing at the same bar
In this thread I wanna have a constructive debate about: operating principle, validity, suggestion, bug and new filter.
Let’s take a close look, No trade filter is perfect, a strategy without trade isn’t a strategy, is nothing.
Too many ambiguos trades, ok here we are this is the first obscure filter, what this really means for you designer/developer? I mean what is ambiguos for you maybe isn’t for me, or viceversa.
No filled trades, nice filter, but i have a question also about this, it consider a strategy that have the last trade (so at the ending of our data) not filled so dismiss this strategy when phreaps it is good?
Outlier trade, statistically correct IMHO.
Zero PL trades and Zero duration trades not need explanation for me, they are mandatory like other that we have just seen.
Unfinished trades, meaning of this filter? Is somehow like No filled trades, isn’t?
Too little trades, what this means? If i’m searching something very fast, like scalping or a fast intraday maybe this filter dismiss all the strategy wich are good for this purpose?
And last one, too many trades closing at the same bar, question, if i’m searching daily strategies and the strategy close the trades on the same daily bar (so a big range of time if you think that someone operate on H1 or less) it will dismissed? if it will is somehow wrong i think.
My Apologies for the big post, but this argument requires explanations, consider this thread like the future charapter of your StrategyQuant X Guide.
As for the automatic filtering you can check this article. Let us know if any questions
You must be logged in to reply to this topic.