I second this.
proceris wrote:
> Maybe an abstract interface? That would allow people to define their
> own projections from market book to a number. Weighted balance and
> simple balance could be two included examples.
>
> On Jun 30, 5:27 am, nonlinear5 <[email protected]> wrote:
>   
>> Up until recently, TWS provided up to 5 levels of market depth, and it
>> made sense to calculate depth balance as the difference between
>> cumulative bid size and cumulative ask size. More precisely, the
>> balance was calculated as
>> balance = 100 * (cumulativeBid - cumulativeAsk) / (cumulativeBid +
>> cumulativeAsk);
>> where cumulativeBid is a simple sum of all bid sizes through all 5
>> levels,
>> and cumulativeAsk is a simple sum of all ask sizes through all 5
>> levels.
>>
>> Now that we have 10 levels of market depth in TWS, the simple
>> summation of sizes is probably not appropriate. That's because the
>> sizes at the top levels (1 to 5) seem more important than the sizes at
>> the bottom levels (6 to 10), and should be weighted more.
>>
>> The question that I ponder is, what kind of weighting should be used?
>> One possible approach is to use a weighting equivalent to the (10 -
>> level). So, the size at level 1 would get a weight of 10, the size at
>> level 2 would get a weight of 9, and so on to level 10 with a weight
>> of 1. This schema would put 10 times more weight on level 1, compared
>> to weight on level 10.
>>
>> If you can think of a better weighting mechanism, please post your
>> thoughts here. Once we settle on the weighting, I'll modify the
>> framework to calculate and record weighted balance, in addition to
>> simple balance.
>>     
>
> >
>
>   


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"JBookTrader" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/jbooktrader?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to