I'm not convinced its a big task because we can treat is as a black box
and use some of the 3rd party implementations. I think it boils down to
figuring out how to apply it to JBT. For example, what is to be
estimated by
the filter ? Perhaps the price reactivity to book indicators ? If so,
the we
need an indicator for "price reactivity to book indicator."
On 10/21/2010 4:02 PM, new_trader wrote:
Perhaps this:http://jkalman.sourceforge.net/
thanks a lot for first feedback!
after a first look at it I have come to the conclusion that this is an
enormous task.
I think the extensive help of the math gurus here in the group is very
important.
from my first observations I would estimate that we need an Unscented
Kalman Filter as the JBT book data seem to be highly non-linear:
http://en.wikipedia.org/wiki/Kalman_filter#Unscented_Kalman_filter
what data sets do we have in JBT:
* we have the book balance showing the situation on the bid and ask
side
* we have the price data showing us the price movements
* we have some pretty cool indicators showing us the "velocity" in
price and book balance movements
Kalman filters seem to be highly used in navigation tasks. Here the
concepts of speed and movement are incorporated into the Kalman
filters to predict movements or to track objects.
What do the math gurus say: is it possible to apply theses concepts
from navigation to the movements of prices?
The (linear) Kalman filter model calculates the next state from a
state transition model F and a control-input model B and the process
noise w and some other factors like an observation model H.
My questions to our math gurus:
* are these models relevant for an Unscented Kalman filter?
* if they are: how will these models be built?
I think the Kalman Filter stuff for JBT is really a very big task!
Please give your help and support so that we can bring something to
fly :-)
--
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.
