Abstract:
The present paper is concerned with the conditionally extremal settings of
the sequential testing problem of two simple hypotheses about a parameter responsible for the local
return rate of a stationary Ornstein–Uhlenbeck
process to its mean value.
Minimization of the Kullback–Leibler divergence is considered as an optimality test.
An asymptotically optimal scheme is put forward, first, in the case when
the error probabilities of the first and the second kind tend to zero,
and, second, in the case when the tested parameters go off to infinity but the distance between them is fixed.
Fulltext:
PDF file (497 kB)