Relationship between spectral and coefficient criteria of mean-square stability for systems of linear stochastic differential and difference equations (Q2722232)

In 1964 I. I. Gikhman obtained a spectral criterion for the mean square stability of the stochastic Itô system NEWLINE\[NEWLINEdx(t)=Ax(t) dt+ Bx(t) dw(t),\tag{1}NEWLINE\]NEWLINE where \(x\in \mathbb R^n, A,B\in \mathbb R^{n\times n}\), and \(w(t)\) is a scalar standard Wiener process (see \textit{I. I. Gikhman} and \textit{A. V. Skorokhod} [Stochastic differential equations and their applications. Kiev: ``Naukova Dumka'' (1982; Zbl 0557.60041)]). The author shows that this criterion is equivalent to the following one: the trivial solution of the system (1) is stable in mean square iff the matrix \(A\) is stable and there exists a positive definite symmetric solution of the matrix algebraic equation \(A^TH+HA+B^THB=-G,\) where \(G\) is an arbitrary symmetric positive definite matrix.NEWLINENEWLINENEWLINESimilar results are obtained for the stochastic difference equation \(x(k+1)=[A +B\xi (k)]x(k)\), \(k=0,1,2,\ldots,\) where \(\xi(k)\) is a standard stationary discrete white noise, as well as for the stochastic functional equation \(x(t+\tau)=[A +B\xi (t)]x(t),\) where \(\xi(t)\) is a stochastic process of white noise type.