Estimation of covariance matrices of vector wiener process by minque method
Suppose \(Y(t)=w(t)+\epsilon (t)\), \(t>T\), is a stochastic process where w is a multidimensional Wiener process with covariance matrix D, i.e. w(t)- w(s)\(\sim N(0,D(t-s))\) and the noise process \(\epsilon\) has orthogonal values and \(\epsilon\) (t)\(\sim N(0,G)\) for each t. It is assumed that w and \(\epsilon\) are independent. The covariance matrices D and G are unknown and the problem is to find their estimators by using observations of the process Y(t) at given times \(t_ 0,t_ 1,...,t_ n\). The estimators \(\hat D\) and \(\hat G\) based on \(Y(t_ 0),Y(t_ 1),...,Y(t_ n)\) are described explicitly. Some arguments are given to illustrate the connection of the above problem with the parametric estimation problem in general linear models.
- Estimation of the matrices of parameters and covariations of the perturbation vectors in multidimensional discrete-time dynamic systems under special structure of the unknown covariance matrices
- On the estimation of structured covariance matrices
- scientific article; zbMATH DE number 4092607
- Maximum likelihood estimators of parameters of multidimensional stationary Gaussian AR processes
- Some estimation problems in infinite dimensional Gaussian white noise
This page was built for publication: Estimation of covariance matrices of vector wiener process by minque method
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3745104)