Effects of statistical dependence on multiple testing under a hidden Markov model - Mathematics > Statistics TheoryReportar como inadecuado




Effects of statistical dependence on multiple testing under a hidden Markov model - Mathematics > Statistics Theory - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Abstract: The performance of multiple hypothesis testing is known to be affected by thestatistical dependence among random variables involved. The mechanismsresponsible for this, however, are not well understood. We study the effects ofthe dependence structure of a finite state hidden Markov model HMM on thelikelihood ratios critical for optimal multiple testing on the hidden states.Various convergence results are obtained for the likelihood ratios as theobservations of the HMM form an increasing long chain. Analytic expansions ofthe first and second order derivatives are obtained for the case of binarystates, explicitly showing the effects of the parameters of the HMM on thelikelihood ratios.



Autor: Zhiyi Chi

Fuente: https://arxiv.org/







Documentos relacionados