Fully Automated Stocks Trading

[samuel11]

Because we don't know for sure that it's winter

GAT

"Une hirondelle ne fait pas le printemps"

 

[Bigchazza]

Et une pandémie ne fait pas l'hiver.

 

[ValeryN]

Feb was +14% month. Coincidentally, nearly identical to Feb last year but with less capital used and smaller positions, which is great.

Some sort of fairy tale. All strategies made money. Timing couldn't have been better to withdraw for last year's taxes.

Funny stats for this month:

Individual strategy + combined (% numbers are slightly off)

Val

 

[Bigchazza]

Hi Val. I'm struggling to interpret your performance chart above. I am certain it's because I don't know what it means, not because it's wrong! My problem is the combined performance in grey bars. Shouldn't it be an average of the individual strategies? E.g. for the first bar - shouldn't it be an avg of 2%, 0.5% and 0.01%? (figures not necessarily accurate - just going by the chart)

 

[ValeryN]

Hi Val. I'm struggling to interpret your performance chart above. I am certain it's because I don't know what it means, not because it's wrong! My problem is the combined performance in grey bars. Shouldn't it be an average of the individual strategies? E.g. for the first bar - shouldn't it be an avg of 2%, 0.5% and 0.01%? (figures not necessarily accurate - just going by the chart)

No worries. Combined is a sum of all strategies.
RealTest offers a nice table view to examine this in details.

Note that there will always be some discrepancy between this view and my live account. Source of diff is mainly

1. Live vs model commissions
2. Live positions carried over from previous month

But normally it is negligible. When I post monthly PnL it's normally from a brokerage statement.

Val

 

[Bigchazza]

Thanks Val. So each of those four blanked out columns above is a strategy? Here is the key question: are they each applied to the same capital simultaneously? E.g. you have a total pot of $1000 and you allocate $1000 to each strategy and let them compete for it? If so, the "Combined" column makes sense to me. If not, and each strategy is allocated e.g. 25% of total capital, it doesn't make sense! Sorry to pick - I find this really interesting.

 

[ValeryN]

Thanks Val. So each of those four blanked out columns above is a strategy? Here is the key question: are they each applied to the same capital simultaneously? E.g. you have a total pot of $1000 and you allocate $1000 to each strategy and let them compete for it? If so, the "Combined" column makes sense to me. If not, and each strategy is allocated e.g. 25% of total capital, it doesn't make sense! Sorry to pick - I find this really interesting.

Yes and yes. One column is one strategy. Each has 100% allocation.

 

[mhparker]

I'm happy to see all those 0s in the discretion column

 

[Bigchazza]

Yes and yes. One column is one strategy. Each has 100% allocation.

I'm getting you now! BTW I took your recommendation and read the Laurens Bendorp book. He talks about being simultaneously 100% long and 100% short, and using leverage to do that. I found the book really helpful but TBH I am still shaky on that side - leverage and the mechanics of combining systems. Got a couple of things in the reading pile that will hopefully fix that.

 

[ValeryN]

He talks about being simultaneously 100% long and 100% short

Not sure if that is what he meant. Artificially maintaining same % invested in both sides will likely significantly reduce your potential returns and might expose you to unnecessary extra risk, unless they have same volatility and positive expectancy, which is unlikely at the same time. But test everything yourself and see what makes sense for you.