A system identification approach to clustering vector autoregressive time series

TL;DR

This paper introduces a novel clustering method for vector autoregressive time series using a system identification approach, addressing computational challenges and providing a scalable solution with effective model selection.

Contribution

It proposes the k-LMVAR algorithm for clustering vector time series based on underlying autoregressive dynamics, overcoming previous computational limitations.

Findings

k-LMVAR performs well in simulations

The method scales efficiently to larger datasets

Effective model order and cluster number selection via BIC

Abstract

Clustering of time series based on their underlying dynamics is keeping attracting researchers due to its impacts on assisting complex system modelling. Most current time series clustering methods handle only scalar time series, treat them as white noise, or rely on domain knowledge for high-quality feature construction, where the autocorrelation pattern/feature is mostly ignored. Instead of relying on heuristic feature/metric construction, the system identification approach allows treating vector time series clustering by explicitly considering their underlying autoregressive dynamics. We first derive a clustering algorithm based on a mixture autoregressive model. Unfortunately it turns out to have significant computational problems. We then derive a `small-noise' limiting version of the algorithm, which we call k-LMVAR (Limiting Mixture Vector AutoRegression), that is computationally manageable. We develop an associated BIC criterion for choosing the number of clusters and model order. The algorithm performs very well in comparative simulations and also scales well computationally.