PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Error exponents in AR order testing
Stephane Boucheron and Elisabeth Gassiat
IEEE Transactions Information Theory Volume 52, Number 3, 2006.

There is a more recent version of this eprint available. Click here to view it.


This paper is concerned with error exponents in testing problems raised by autoregressive (\textsc{ar}) modeling. The tests to be considered are variants of generalized likelihood ratio testing corresponding to traditional approaches to auto-regressive moving-average (ARMA) modeling estimation. In several related problems like Markov order or hidden Markov model order estimation, optimal error exponents have been determined thanks to large deviations theory. \textsc{ar} order testing is specially challenging since the natural tests rely on quadratic forms of Gaussian processes. In sharp contrast with empirical measures of Markov chains, the large deviation principles satisfied by Gaussian quadratic forms do not always admit an information-theoretical representation. Despite this impediment, we prove the existence of non-trivial error exponents for Gaussian \textsc{ar} order testing. And furthermore, we exhibit situations where the exponents are optimal. These results are obtained by showing that the log-likelihood process indexed by \textsc{ar} models of a given order satisfy a large deviation principle upper-bound with a weakened information-theoretical representation.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:557
Deposited By:Stephane Boucheron
Deposited On:25 December 2004

Available Versions of this Item