ON INTEGRATED SQUARE ERRORS OF RECURSIVE NONPARAMETRIC
ESTIMATES OF NONSTATIONARY MARKOV PROCESSES
Abstract: The integrated square error (ISE) and the mean integrated square error
(MISE) for a class of recursive estimators of the transition density function of a
vector-valued nonstationary Markov process are considered. Conditions are given under
which the MISE converges, and the ISE converges in probability and almost surely.
2000 AMS Mathematics Subject Classification: Primary: -; Secondary: -;
Key words and phrases: -