Sample-complexity bound for high-order linear systems
Produced by Long-Context Linear System Identification
Vector autoregressive process
A multivariate stochastic process in which each vector-valued state is a linear combination of a fixed number of preceding states plus an innovation term.
A multivariate stochastic process in which each vector-valued state is a linear combination of a fixed number of preceding states plus an innovation term.
In the long-context system-identification setting, a VAR(\(p\)) process is the observed-state formulation of the high-order linear dynamical system under study. It can also be written as a first-order linear system by stacking the previous \(p\) states.