Recursive estimation of the transition distribution function of a Markov process: Asymptotic normality (Q1176989)

Consider a strictly stationary Markov process \(X_ 1,X_ 2,\ldots.\) A recursive kernel-based nonparametric estimator of the one-step transition distribution is shown to be asymptotically normal, under stated regularity conditions. The class of Markov processes satisfying these conditions includes the Markov processes usually considered in the literature; namely, processes which either satisfy Doeblin's hypothesis, or, more generally, are geometrically ergodic. [Editorial remark: See also the author's article reviewed above.].