VARX-L: Structured regularization for large vector autoregressions with exogenous variables

The vector autoregression (VAR) has long proven to be an effective method for modeling the joint dynamics of macroeconomic time series, as well as for forecasting. One major shortcoming of the VAR that has limited its applicability is its heavy parameterization: the parameter space grows quadratically with the number of series included, quickly exhausting the available degrees of freedom. Consequently, using VARs for forecasting is intractable for low-frequency, high -dimensional macroeconomic data. However, empirical evidence suggests that VARs that incorporate more component series tend to result in more accurate forecasts. Most conventional methods that allow for the estimation of large VARs either require ad hoc subjective specifications or are computationally infeasible. Moreover, as global economies become more intricately intertwined, there has been a substantial interest in incorporating the impact of stochastic, unmodeled exogenous variables. Vector autoregression with exogenous variables (VARX) extends the VAR to allow for the inclusion of unmodeled variables, but faces similar dimensionality challenges. This paper introduces the VARX-L framework, a structured family of VARX models, and provides a methodology that allows for both efficient estimation and accurate forecasting in high -dimensional analysis. VARX-L adapts several prominent scalar regression regularization techniques to a vector time series context, which greatly reduces the parameter space of VAR and VARX models. We also highlight a compelling extension that allows for shrinking toward reference models, such as a vector random walk. We demonstrate the efficacy of VARX-L in both low-and high -dimensional macroeconomic forecasting applications and simulated data examples. Our methodology is easy to reproduce in a publicly available Rpackage. (C) 2017 International Institute of Forecasters. Published by Elsevier B.V. All rights reserved.