Bayesian doubly adaptive elastic-net Lasso for VAR shrinkage

Abstract

We develop a novel Bayesian doubly adaptive elastic-net Lasso (DAELasso) approach for VAR shrinkage. DAELasso achieves variable selection and coefficients shrinkage in a data based manner. It constructively deals with the explanatory variables that tend to be highly collinear by encouraging grouping effect. In addition, it allows for different degree of shrinkages for different coefficients. Rewriting the multivariate Laplace distribution as a scale mixture, we establish closed-form conditional posteriors that can be drawn from a Gibbs sampler. Empirical analysis shows that forecast results produced by DAELasso and its variants are comparable to that of other popular Bayesian methods, which provides further evidence that the forecast performances of large and medium sized Bayesian VARs are relatively robust to prior choices, and in practice simple Minnesota types of priors can be more attractive relative to their complex and well designed alternatives. * I would like to thank Gary Koop, Esther Ruiz and two anonymous referees for their constructive comments. I would also like to thank the conference participants of CFE11, ESEM2012, and RCEF2012 for helpful discussions. Any remaining errors are my own responsibility.

Adaptive Shrinkage in Bayesian Vector Autoregressive Models

“…Gefang (2014) propose a Bayesian LASSO for VAR models, showing that forecast results tend to be similar to the ones obtained from standard Minnesota-type priors. However, it is worth noting that a major limitation of the double exponential prior is its lack of flexibility.…”

The shortage of safe assets in the US investment portfolio: Some international evidence

“…In the empirical application we set κ ξ0 = κ ξ1 = 0.01 to induce large amounts of shrinkage on all coefficients of a given country and set ϑ τ i = 0.6 for all countries. Note that ϑ τ i = 1 would lead to the Bayesian LASSO (see, for example, Kozumi and Kobayashi, 2011;Gefang, 2014).…”

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

“…Theoretical properties were investigated by Kock and Callot (2015) and by . Gefang (2012) considers a Bayesian implementation of the elastic net, an extension of the lasso proposed by Zou and Hastie (2005) that accounts for highly correlated covariates. However, their implementation is not computationally tractable and they do not observe much of a forecasting improvement over existing methods.…”