SLANTS: Sequential Adaptive Nonlinear Modeling of Vector Time Series

TL;DR

This paper introduces SLANTS, an adaptive online method for nonlinear vector time series modeling using spline basis expansion and l1-regularization, with proven error bounds and variable selection consistency.

Contribution

It presents a novel adaptive filtering algorithm that models nonlinear vector time series with spline basis and regularization, enabling automatic parameter tuning and theoretical performance guarantees.

Findings

Effective modeling of synthetic and real-world data.

Theoretical bounds on prediction errors.

Proven variable selection consistency.

Abstract

We propose a method for adaptive nonlinear sequential modeling of vector-time series data. Data is modeled as a nonlinear function of past values corrupted by noise, and the underlying non-linear function is assumed to be approximately expandable in a spline basis. We cast the modeling of data as finding a good fit representation in the linear span of multi-dimensional spline basis, and use a variant of l1-penalty regularization in order to reduce the dimensionality of representation. Using adaptive filtering techniques, we design our online algorithm to automatically tune the underlying parameters based on the minimization of the regularized sequential prediction error. We demonstrate the generality and flexibility of the proposed approach on both synthetic and real-world datasets. Moreover, we analytically investigate the performance of our algorithm by obtaining both bounds of the prediction errors, and consistency results for variable selection.