Sequential estimation for continuous time finite markov processes

From MaRDI portal





The problem of efficient sequential estimation in finite-state Markov processes is considered. Probably the authors have not seen the paper of \textit{S. Trybuła}, Zastosow. Mat. 17, 227-248 (1982; Zbl 0514.62094), where all their results concerning the information inequality and efficient sequential estimation of a single row of the intensity matrix have been obtained. The authors treat also briefly the problem of efficient sequential estimation for functions depending on more than one row of the intensity matrix. However their argument that an efficiently estimable function of two parameters (say \(g(\theta_ 1,\theta_ 2))\) remains efficiently estimable as a function of one parameter (say \(g(\theta_ 1,\theta_ 2)\) with \(\theta_ 2\) fixed), is not true. The latter implies also incompleteness of section 6 of the paper.











This page was built for publication: Sequential estimation for continuous time finite markov processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3223739)