Numerical bounds on fluctuating linear processes
From MaRDI portal
Let \(A\) be an \(n\times n\) primitive nonnegative matrix. The long-run behavior \(x_k/\|x_k\|_1\) of the linear process \(x_{k+1}=x_k A\) is determined by the stochastic eigenvector \(\pi\) of \(A\). The author considers the linear process \(x_{k+1}=x_k A_k\), where each \(A_k\) fluctuates about \(A\), and provides numerical bounds on the difference between \(x_k/\|x_k\|\) and \(\pi\), thus showing how well \(\pi\) describes the long-run behavior of this fluctuation behavior.
Recommendations
- On the Dynamics of the Linear Process $Y( k ) = A ( k )Y ( k - 1 )$ with Irreducible Matrices $A( k )$
- On the growth of infinite products of slowly varying primitive matrices
- A note on ergodicity for nonautonomous linear difference equations
- Upper bounds on the maximum modulus of subdominant eigenvalues of nonnegative matrices
- Asymptotic properties of powers of nonnegative matrices, with applications
Cites work
This page was built for publication: Numerical bounds on fluctuating linear processes
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5927713)
