Non-parametric State Space Models
A.V. Dobrovidov, G.M. Koshkin, and V.A. Vasiliev
Press, 2012, ii + 501pp.
A key problem of mathematical statistics is that of reconstructing
probability distributions of random variables and functionals of
these distributions, such as moment characteristics and regression
function. In the present volume the approach is non-parametric;
knowledge of a parametric structure for the class of estimated
functionals is not assumed. It is supposed that observations are
statistically dependent, which permits the use of dynamical
systems in constructing models. The book will be useful for
experts in information processing, control, prediction and
filtration, and for students specializing in mathematical statistics.
To evaluate whether the book will be useful to you, please examine the Table of Contents*.
*To view the contents pages will require Adobe Reader*