Thanks for that link - looks like there was considerable interest and its been 
planned since 2006.  Anyone know why it wasn't ultimately implemented?  

 

________________________________
From: "[email protected]" <[email protected]>
To: [email protected]
Sent: Wed, February 10, 2010 9:07:09 AM
Subject: [amibroker] Re: Multiple Systems & Risk Mgmt in one formula

  
Getting extra data into the CBT can be a challenge. There is a suggestion in 
the feedback center here:
http://www.amibroke r.com/feedback/ view_bug. php?bug_id= 598

--- In amibro...@yahoogrou ps.com, B S <bs2...@...> wrote:
>
> Hi-
> 
> I asked Marcin a question related to this earlier, but given the considerable 
> interest in the past on this board regarding managing multiple systems, I'd 
> thought put some further thoughts here.  My understanding of the conclusion 
> reached during the last round of discussions was that each individual should 
> use low-level CBT to accomplish their multiple systems management goals.  So 
> I've done that, but in the end its of no practical use because of the 
> incredible number of static arrays that were required to do it - it 'works' 
> on small samples/tests but the memory demands are too great for things that 
> I'd actually use it for.  As my objectives were rather simple, I imagine that 
> others are still wrestling with similar issues.  
> 
> As an example, part of my plan was to limit the % of equity that was employed 
> by any one system at a given time.  Therefore, while in the backtester I 
> needed to know which system or systems generated the signal.  So I started by 
> creating a static array that i could later access in the CBT which 
> contained flags (1,2,4,etc.) indicating which system or systems it came 
> from.  However, because of the memory issues, I decided to stuff these flags 
> into the positionsize property and just determine the actual size later while 
> looping through CBT.  That allowed me to determine which systems signals were 
> coming from, and then choose which model to allocate the trade to (in the 
> event of multiple simultaneous signals), but I still needed to get the 
> correct entry price.  There could be as many entry prices as I have systems, 
> and since I don't know which I'll be using until I get to the backtester, I 
> needed to store all of them  - another static
> array.  This process continued like this and then it occurred to me that just 
> about all of these issues would go away if I could just attach the 
> information I needed to the entry signal object.  So for example, the user 
> could set PositionSize1, PositionSize2, PositionScore1, PositionScore2,  
> BuyPrice1, BuyPrice2, and so on...or perhaps just dummy variables... Dummy1, 
> Dummy2, etc..   I'm sure there's a memory cost to doing this, but it can't be 
> anything like storing all these static arrays (is this correct? i really have 
> no idea what i'm talking about).  Also, I'm sure just making these additional 
> property slots available will come at some performance cost, but perhaps 
> adding them could be an option in settings.  
> 
> For all I know, editing the number of properties that can be stored in a 
> signal object is already possible - is it?  If not, do others think this 
> would be useful?  Is there a better way of achieving the types of things that 
> I'm trying to do?  
> 
> Appreciate any dialogue whatsoever on this - have spent an embarassing amount 
> of time on it without much to show for it.
>





      

Reply via email to