Factor Modelling for Biclustering Large-dimensional Matrix-valued Time Series

TL;DR

This paper introduces a new unsupervised biclustering method for large matrix-valued time series using a novel two-way factor structure, with theoretical guarantees and practical validation.

Contribution

It proposes a new latent two-way factor model for biclustering, along with a $K$-means algorithm and eigenvalue-ratio method, with improved convergence rates and consistency results.

Findings

Faster convergence rates for global loading matrices.

Consistent estimation of factor numbers and cluster memberships.

Validated with simulated and real data.

Abstract

A novel unsupervised learning method is proposed in this paper for biclustering large-dimensional matrix-valued time series based on an entirely new latent two-way factor structure. Each block cluster is characterized by its own row and column cluster-specific factors in addition to some common matrix factors which impact on all the matrix time series. We first estimate the global loading spaces by projecting the observation matrices onto the row or column loading space corresponding to common factors. The loading spaces for cluster-specific factors are then further recovered by projecting the observation matrices onto the orthogonal complement space of the estimated global loading spaces. To identify the latent row/column clusters simultaneously for matrix-valued time series, we provide a $K$-means algorithm based on the estimated row/column factor loadings of the cluster-specific weak factors. Theoretically, we derive faster convergence rates for global loading matrices than those of the state-of-the-art methods available in the literature under mild conditions. We also propose an one-pass eigenvalue-ratio method to estimate the numbers of global and cluster-specific factors. The consistency with explicit convergence rates is also established for the estimators of the local loading matrices, the factor numbers and the latent cluster memberships. Numerical experiments with both simulated data as well as a real data example are also reported to illustrate the usefulness of our proposed method.