Hi,

thanks for the important comments and clarification.
I am now wondering whether you all see the same market depth as I do..
If I use market depth tool by IB for EUR-Future I see typically 6 bid
levels, 5 ask levels.
No more.
Probably this is the reason why it is considered invalid.

Regarding the issue: I see it such that currently market validation
bans some instruments
from being traded. So, I see two possibilities (if this should be
repaired):
a) market depth that is taken into account, could be a per-contract
parameter.
   (meaning, one would only look at the first 5 levels for EUR, but
for the first 10 levels for ES)
b) invalid data is simply passed along with a tag "invalid". Then each
strategy can decide whether
   after a number of ticks it simply "believes" the data.

c) a combination of both.

Certainly, the current situation where data stalls for invalid
received data is problematic. This means
that most data on contracts like EUR-future will not be accepted (not
just a few gaps in between? - Most would be gaps..
Right?
(At least, here most! of the data on record is identical..)

Klaus


On 4 Aug., 14:31, Eugene Kononov <[email protected]> wrote:
> I can make the market depth validation optional via the user preference, but
> this is likely to lead to unreliable results.
>
> What can be thought as a "true" market depth is a continuously moving
> target. For liquid instruments such as the ES during the regular trading
> session, JBT gets about 50 messages per second from the IB servers
> indicating changes in market depth. Now, consider what happens when the
> market moves. Suppose the best bid for the ES moves up by 1 tick from 900.25
> to 900.50. The way the IB server communicates this change might be these
> series of updates:
>
> -- delete bid level 1 from the queue
> -- shift the remaining levels 2 through 9 up by 1
> -- insert level 10
>
> All of these are separate messages, delivered within a few milliseconds.
> Now, if we happen to take a snapshot of market depth immediately after the
> "delete" message, we would likely find that the cumulative size of the bids
> has suddenly dropped by about 10%. Is that representative of the current
> balance of buyers and sellers? Of course not! This is simply the result of
> the mechanics of the update which removes the head of the queue and makes
> more of its tail visible. At the end of the update, the number of bid levels
> will be exactly the same as the number of ask levels, and the cumulative
> size of all the bids will probably be about the same as it was before the
> update.
>
> Now, I understand that there are some less liquid instruments which may
> legitimately have unequal number of bid and ask levels for some prolonged
> time. However, If there is no validation, what would happen is that JBT
> would sometimes catch market depth in "transit", in an unfinished state, as
> described above.  With these less liquid instruments, it becomes more
> difficult to determine whether the unequal number of bid/ask levels is the
> result of the on-going market depth update, or the true representation of
> the buy/sell orders.
--~--~---------~--~----~------------~-------~--~----~
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