Constrained hypothesis testing and the Cramér-Rao bound

The classical Wald and Rao test statistics are asymptotically equivalent to the generalized likelihood ratio test statistics, while not requiring parameter estimation under both hypotheses, and so they provide lower complexity test statistics. In this paper we develop corresponding variations of the Wald and Rao test for nested hypothesis testing under parameter constraints. The resulting tests incorporate the constrained Cramér-Rao bound formulation from Stoica and Ng, and unify some asymptotic hypothesis testing results. Examples will illustrate key ideas and test performance.

[1]  GourierouxMonfort.  Statistics and Econometric Models, Volume 2 , 1996 .

[2]  T. Rothenberg.  Identification in Parametric Models , 1971 .

[3]  C. G. Khatri,et al.  A note on a manova model applied to problems in growth curve , 1966 .

[4]  Eric Moulines,et al.  Subspace methods for the blind identification of multichannel FIR filters , 1995, IEEE Trans. Signal Process..

[5]  S. D. Silvey,et al.  The Lagrangian Multiplier Test , 1959 .

[6]  Alain Monfort,et al.  General concepts, estimation, prediction, and algorithms , 1995 .

[7]  Alfred O. Hero,et al.  Lower bounds for parametric estimation with constraints , 1990, IEEE Trans. Inf. Theory.

[8]  B. C. Ng,et al.  On the Cramer-Rao bound under parametric constraints , 1998, IEEE Signal Processing Letters.

[9]  D. Sattinger,et al.  Calculus on Manifolds , 1986 .

[10]  John Aitchison,et al.  Large‐Sample Restricted Parametric Tests , 1962 .

[11]  R. F.,et al.  Mathematical Statistics , 1944, Nature.

[12]  J. Munkres,et al.  Calculus on Manifolds , 1965 .

[13]  Adi Ben-Israel,et al.  Generalized inverses: theory and applications , 1974 .

[14]  M. Crowder.  On constrained maximum likelihood estimation with non-i.i.d. observations , 1984 .

[15]  J. Hosking.  Lagrange Multiplier Test , 2006 .

[16]  Alle-Jan van der Veen,et al.  Asymptotic properties of the algebraic constant modulus algorithm , 2001, IEEE Trans. Signal Process..

[17]  Brian M. Sadler,et al.  The Constrained CramÉr–Rao Bound From the Perspective of Fitting a Model , 2007, IEEE Signal Processing Letters.

[18]  Brian M. Sadler,et al.  Maximum-Likelihood Estimation, the CramÉr–Rao Bound, and the Method of Scoring With Parameter Constraints , 2008, IEEE Transactions on Signal Processing.