On Multivariate Singular Spectrum Analysis

by   Anish Agarwal, et al.

We analyze a variant of multivariate singular spectrum analysis (mSSA), a widely used multivariate time series method, which we find to perform competitively with respect to the state-of-art neural network time series methods (LSTM, DeepAR). Its restriction for single time series, singular spectrum analysis (SSA), has been analyzed recently. Despite its popularity, theoretical understanding of mSSA is absent. Towards this, we introduce a natural spatio-temporal factor model to analyze mSSA. We establish the in-sample prediction error for imputation and forecasting under mSSA scales as 1/√(NT), for N time series with T observations per time series. In contrast, for SSA the error scales as 1/√(T) and for matrix factorization based time series methods, the error scales as 1/min(N, T). We utilize an online learning framework to analyze the one-step-ahead prediction error of mSSA and establish it has a regret of 1/(√(N)T^0.04) with respect to in-sample forecasting error. By applying mSSA on the square of the time series observations, we furnish an algorithm to estimate the time-varying variance of a time series and establish it has in-sample imputation / forecasting error scaling as 1/√(NT). To establish our results, we make three technical contributions. First, we establish that the "stacked" Page Matrix time series representation, the core data structure in mSSA, has an approximate low-rank structure for a large class of time series models used in practice under the spatio-temporal factor model. Second, we extend the theory of online convex optimization to address the variant when the constraints are time-varying. Third, we extend the analysis prediction error analysis of Principle Component Regression beyond recent work to when the covariate matrix is approximately low-rank.


page 1

page 2

page 3

page 4


Matrix factorization for multivariate time series analysis

Matrix factorization is a powerful data analysis tool. It has been used ...

Nonstationary Temporal Matrix Factorization for Multivariate Time Series Forecasting

Modern time series datasets are often high-dimensional, incomplete/spars...

High-Dimensional Multivariate Forecasting with Low-Rank Gaussian Copula Processes

Predicting the dependencies between observations from multiple time seri...

Compact representation of temporal processes in echosounder time series via matrix decomposition

Echosounders are high-frequency sonar systems widely used to observe mid...

Network Autoregression for Incomplete Matrix-Valued Time Series

We study the dynamics of matrix-valued time series with observed network...

SAMoSSA: Multivariate Singular Spectrum Analysis with Stochastic Autoregressive Noise

The well-established practice of time series analysis involves estimatin...

Modeling Interval Trendlines: Symbolic Singular Spectrum Analysis for Interval Time Series

In this article we propose an extension of singular spectrum analysis fo...

Please sign up or login with your details

Forgot password? Click here to reset