Combining shrinkage and sparsity in conjugate vector autoregressive models

Hauzenberger, Niko; Huber, Florian; Onorante, Luca

doi:10.1002/jae.2807

Cited by 10 publications

(7 citation statements)

References 57 publications

(106 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…( 5), the SAVS estimator can be interpreted as special case of the coordinate descent algorithm (Friedman et al, 2007) by relying on a single iteration to obtain a closed-form solution. Ray and Bhattacharya (2018) and Hauzenberger et al (2020b) both provide evidence that the coordinate descent algorithm already converges after the first pass through.…”

Section: Sparsifying the Cointegration Matrixmentioning

confidence: 92%

“…(2020), Ray andBhattacharya (2018), andBashir et al (2019), which have been successfully used in a range of multivariate and univariate macroeconomic and finance applications (see Puelz et al, 2017;Huber et al, 2020a;Hauzenberger et al, 2020b). In cointegration models with time-varying parameters, however, it is necessary to adjust the proposed procedures for specific parts of the parameter space.…”

Section: Dynamic Sparsificationmentioning

confidence: 99%

“…( 5) can be interpreted as minimizing the expected loss (see Hahn and Carvalho, 2015). By following Huber et al (2020a) and Hauzenberger et al (2020b) we deviate from this procedure and solve the optimization problem for each draw from the posterior distribution. Setting Πt = Π (s) t with (s) indicating the sth draw, has the convenient feature of effectively allowing for uncertainty quantification of the rank of Π t .…”

Section: Sparsifying the Cointegration Matrixmentioning

confidence: 99%

See 2 more Smart Citations

Sparse time-varying parameter VECMs with an application to modeling electricity prices

Hauzenberger¹,

Pfarrhofer²,

Rossini³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

In this paper we propose a time-varying parameter (TVP) vector error correction model (VECM) with heteroscedastic disturbances. We combine a set of econometric techniques for dynamic model specification in an automatic fashion. We employ continuous global-local shrinkage priors for pushing the parameter space towards sparsity. In a second step, we post-process the cointegration relationships, the autoregressive coefficients and the covariance matrix via minimizing Lasso-type loss functions to obtain truly sparse estimates. This two-step approach alleviates overfitting concerns and reduces parameter estimation uncertainty, while providing estimates for the number of cointegrating relationships that varies over time. Our proposed econometric framework is applied to modeling European electricity prices and shows gains in forecast performance against a set of established benchmark models.

show abstract

Section: Sparsifying the Cointegration Matrixmentioning

confidence: 92%

Section: Dynamic Sparsificationmentioning

confidence: 99%

Section: Sparsifying the Cointegration Matrixmentioning

confidence: 99%

See 1 more Smart Citation

Sparse time-varying parameter VECMs with an application to modeling electricity prices

Hauzenberger¹,

Pfarrhofer²,

Rossini³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…These typically involve the use of principal components methods to extract the information in the large number of variables into a small number of factors thus avoiding over-parameterization concerns. Starting with Banbura, Giannone, and Reichlin (2010) many researchers have been simply including all the variables in a Vector Autoregression (VAR) and using Bayesian shrinkage priors to avoid over-fitting (see, among many others, Koop (2013), Giannone, Lenza, and Primiceri (2015), Jarocinski and Mackowiak (2017), Carriero, Clark, and Marcellino (2019), Koop and Korobilis (2019), Korobilis and Pettenuzzo (2019), Giannone, Lenza, and Primiceri (2019), Huber and Feldkircher (2019), Chan (2020) and Hauzenberger, Huber, and Onorante (2021)).…”

Section: Introductionmentioning

confidence: 99%

Subspace Shrinkage in Conjugate Bayesian Vector Autoregressions

Huber¹,

Koop²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Macroeconomists using large datasets often face the choice of working with either a large Vector Autoregression (VAR) or a factor model. In this paper, we develop methods for combining the two using a subspace shrinkage prior. Subspace priors shrink towards a class of functions rather than directly forcing the parameters of a model towards some pre-specified location. We develop a conjugate VAR prior which shrinks towards the subspace which is defined by a factor model. Our approach allows for estimating the strength of the shrinkage as well as the number of factors. After establishing the theoretical properties of our proposed prior, we carry out simulations and apply it to US macroeconomic data. Using simulations we show that our framework successfully detects the number of factors. In a forecasting exercise involving a large macroeconomic data set we find that combining VARs with factor models using our prior can lead to forecast improvements.

show abstract

“…Huber et al (2021) andHauzenberger et al (2021) apply SAVS to multivariate time series models and show that it works well for forecasting.…”

mentioning

confidence: 99%

Nonlinearities in Macroeconomic Tail Risk through the Lens of Big Data Quantile Regressions

Prüser¹,

Huber²

2023

Preprint

View full text Add to dashboard Cite

Modeling and predicting extreme movements in GDP is notoriously difficult and the selection of appropriate covariates and/or possible forms of nonlinearities are key in obtaining precise forecasts. In this paper, our focus is on using large datasets in quantile regression models to forecast the conditional distribution of US GDP growth. To capture possible non-linearities we include several nonlinear specifications. The resulting models will be huge dimensional and we thus rely on a set of shrinkage priors. Since Markov Chain Monte Carlo estimation becomes slow in these dimensions, we rely on fast variational Bayes approximations to the posterior distribution of the coefficients and the latent states. We find that our proposed set of models produces precise forecasts. These gains are especially pronounced in the tails. Using Gaussian processes to approximate the nonlinear component of the model further improves the good performance in the tails.

show abstract

Combining shrinkage and sparsity in conjugate vector autoregressive models

Cited by 10 publications

References 57 publications

Sparse time-varying parameter VECMs with an application to modeling electricity prices

Sparse time-varying parameter VECMs with an application to modeling electricity prices

Subspace Shrinkage in Conjugate Bayesian Vector Autoregressions

Nonlinearities in Macroeconomic Tail Risk through the Lens of Big Data Quantile Regressions

Contact Info

Product

Resources

About