Tail probabilities for the null distribution of scanning statistics

This paper is concerned with statistics that scan a multidimensional spatial region to detect a signal against a noisy background. The background is modelled as independent observations from an exponential family of distributions with a known 'null' value of the natural parameter, while the signal is given by independent observations from the same exponential family, but with a different value of the parameter on a particular subregion of the spatial domain. The main result is an extension to multidimensional time of the method of Pollak and Yakir, which relies on a change of measure motivated by change-point analysis, to evaluate approximately the null distribution of the likelihood ratio statistic. Both large-deviation and Poisson approximations are obtained.

[1]  Michael Woodroofe,et al.  Frequentist properties of Bayesian sequential tests , 1976 .

[2]  J. Pickands.  Upcrossing probabilities for stationary Gaussian processes , 1969 .

[3]  Amir Dembo,et al.  Statistical Composition of High-Scoring Segments from Molecular Sequences , 1990 .

[4]  D. Siegmund,et al.  Large deviations for the maxima of some random fields , 1986 .

[5]  D. Siegmund,et al.  Tail approximations for maxima of random fields , 1992 .

[6]  C. Loader.  Large-deviation approximations to the distribution of scan statistics , 1991, Advances in Applied Probability.

[7]  D. Aldous.  Probability Approximations via the Poisson Clumping Heuristic , 1988 .

[8]  Hisao Watanabe,et al.  Asymptotic properties of Gaussian random fields , 1973 .

[9]  William Feller,et al.  An Introduction to Probability Theory and Its Applications , 1951 .

[10]  P. A. P. Moran,et al.  An introduction to probability theory , 1968 .

[11]  Daniel Rabinowitz,et al.  Detecting clusters in disease incidence , 1994 .

[12]  David Siegmund,et al.  Approximate Tail Probabilities for the Maxima of Some Random Fields , 1988 .

[13]  A new representation for a renewal-theoretic constant appearing in asymptotic approximations of large deviations , 1998 .

[14]  D. Siegmund,et al.  Using the Generalized Likelihood Ratio Statistic for Sequential Detection of a Change-Point , 1995 .

[15]  M. Woodroofe.  Nonlinear Renewal Theory in Sequential Analysis , 1987 .

[16]  L. Gordon,et al.  Two moments su ce for Poisson approx-imations: the Chen-Stein method , 1989 .

[17]  D. Siegmund.  Sequential Analysis: Tests and Confidence Intervals , 1985 .

[18]  P. Bickel,et al.  Two-Dimensional Random Fields , 1973 .

[19]  J. Kline,et al.  The cusum test of homogeneity with an application in spontaneous abortion epidemiology. , 1985, Statistics in medicine.