Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
204 IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 51, NO. 1, JANUARY 2003 Partial Likelihood for Signal Processing
 

Summary: 204 IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 51, NO. 1, JANUARY 2003
Partial Likelihood for Signal Processing
Tülay Adali, Senior Member, IEEE, and Hongmei Ni
Abstract--We present partial likelihood (PL) as an effective
means for developing nonlinear techniques for signal processing.
Posing signal processing problems in a likelihood setting provides
a number of advantages, such as allowing the use of powerful tools
in statistics and easy incorporation of model order/complexity
selection into the problem by use of appropriate information-theo-
retic criteria. However, likelihood formulations in most time series
applications require a mechanism to discount the dependence
structure of the data. We address how PL bypasses this require-
ment and note that it might coincide with conditional likelihood
in a number of cases. We show that PL theory can also be used
to establish the fundamental information-theoretic connection,
to show the equivalence of likelihood maximization and relative
entropy minimization without making the assumption of indepen-
dent observations, which is an unrealistic assumption for most
signal processing applications. We show that this equivalence is
true for the basic class of probability models (the exponential

  

Source: Adali, Tulay - Department of Computer Science and Electrical Engineering, University of Maryland, Baltimore County

 

Collections: Engineering