New paper on stochastic time series modeling

[Herbert Jaeger]

Dear Connectionists, 

I would like to announce the paper, 

Herbert Jaeger, "Observable operator models for discrete stochastic
time series", accepted for publication by Neural Computation

Abstract:

A widely used class of models for stochastic 
systems is Hidden Markov models. Systems which can be modeled 
by hidden Markov models are a proper subclass of 
*linearly dependent processes*, a class of stochastic 
systems known from mathematical investigations carried out 
over the last four decades. This article provides a novel, 
simple characterization of linearly dependent processes, 
called observable operator models. The mathematical 
properties of observable operator models lead to a 
constructive learning algorithm for the identification of 
linearly dependent processes. The core of the algorithm has a 
time complexity of O(N + n m^3), where N is the size of 
training data, n is the number of  distinguishable outcomes of 
observations, and m is model state space dimension.


A preprint of the paper is available electronically at

ftp://ftp.gmd.de/GMD/ais/publications/1999/jaeger.99.neco.pdf
(PDF format, 410 K)
ftp://ftp.gmd.de/GMD/ais/publications/1999/jaeger.99.neco.ps.gz
(g'zipped PostScript format, 700 K)

I warmly appreciate your comments!

Sincerely, Herbert Jaeger


----------------------------------------------------------------
Dr. Herbert Jaeger                  Phone  +49-2241-14-2253
German National Research Center     Fax    +49-2241-14-2384
for Information Technology (GMD)    email  herbert.jaeger at gmd.de
AiS.BE 
Schloss Birlinghoven                
D-53754 Sankt Augustin, Germany


http://www.gmd.de/People/Herbert.Jaeger/