Guaranteed stability of recursive multi-input-single-output time series models

Recursive time series models can describe effectively and accurately complex systems with time-varying parameters. These simple models can be used in forecasting and control systems. However, these models may be unstable because of plant and measurement noise even when the process is known to be stable. In this paper, we propose an approach to guarantee the stability of time series models by using the Gershgorin Circle Theorem. Data from real patients with Type 1 Diabetes are used to illustrate the performance of the proposed approach. Results show that the proposed method provides stable models. The method can be easily implemented to single- or multi-input-output time series modeling and subspace identification.