Inducing Sparsity and Shrinkage in Time-Varying Parameter Models

Huber, Florian; Koop, Gary; Onorante, Luca

doi:10.1080/07350015.2020.1713796

Cited by 65 publications

(66 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(2020) model the time‐varying regressison coefficients using a reduced‐rank structure, and Huber et al . (2020) develop a method that first shrinks the time‐varying coefficients, followed by setting the small values to zero.…”

Section: Extensions and New Dma Modelsmentioning

confidence: 99%

An Overview of Dynamic Model Averaging Techniques in Time‐series Econometrics

Nonejad¹

2021

Journal of Economic Surveys

View full text Add to dashboard Cite

Dynamic model averaging (DMA) has become a widely used estimation technique in macroeconomic applications. Since its introduction in econom(etr)ics by Gary Koop and Dimitris Korobilis in 2009, applications of DMA have increased in unimaginable ways. Besides applying the original (univariate) framework suggested by Koop and Korobilis on the data of interest, for example, the inflation rate of the country of choice or return on the rate of equity, practitioners have been able to use DMA-based techniques to extend current models, thereby further improving out-of-sample forecast accuracy, overcome computational bottlenecks, and even help improve our understanding of economic phenomena by introducing new models. These include using Google search data in combination with the predictive likelihood to govern switching between different predictive regressions in the model set or specifying large time-varying parameter vector autoregressions that can be estimated without resorting to simulation-based techniques. This study provides an overview of DMA techniques and the ways in which they have evolved since the contribution of Koop and Korobilis.

show abstract

Section: Extensions and New Dma Modelsmentioning

confidence: 99%

An Overview of Dynamic Model Averaging Techniques in Time‐series Econometrics

Nonejad¹

2021

Journal of Economic Surveys

View full text Add to dashboard Cite

show abstract

“…One could ask whether it makes a big difference to zero out different a i 's as opposed to setting them close to zero. Setting them close but not exactly to zero essentially implies that there exists a lower bound of accuracy one can achieve under the specific prior distribution (Huber et al, 2020). For small‐scale systems, this has negligible implications on predictive accuracy.…”

Section: Achieving Sparsity In Var Modelsmentioning

confidence: 99%

“…Apart from reduced flexibility in terms of covariate selection across equations, typical shrinkage priors push many VAR coefficients toward zero. Under continuous shrinkage priors, however, this implies that the probability of observing a coefficient that exactly equals zero is zero (see, e.g., Bhattacharya, Pati, Pillai, & Dunson, 2015; Carvalho, Polson, & Scott, 2010; Griffin & Brown, 2010; Huber & Feldkircher, 2019; Huber, Koop, & Onorante, 2020). Spike and slab priors allow for shrinking coefficients exactly to zero.…”

Section: Introductionmentioning

confidence: 99%

“…But instead of performing cross validation, this task depends only on a single tuning parameter that needs to be chosen by the researcher. These techniques that combine shrinkage and sparsity have been shown to work well in a wide range of applications ranging from finance (Puelz, Hahn, & Carvalho, 2017; 2020) to macroeconomics (Huber et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

“…Our model allows for different covariates across equations, sparsity in terms of the VAR coefficients, and data‐based zero restrictions on the covariance parameters in Σ . Instead of post‐processing posterior mean/median estimates, we follow Huber et al (2020) in sparsifying each draw from the posterior distribution. This yields an approximate posterior distribution for a sparse vector of coefficients that can be used for uncertainty quantification.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Combining shrinkage and sparsity in conjugate vector autoregressive models

Hauzenberger

Huber

Onorante

2021

J of Applied Econometrics

Self Cite

View full text Add to dashboard Cite

Summary Conjugate priors allow for fast inference in large dimensional vector autoregressive (VAR) models. But at the same time, they introduce the restriction that each equation features the same set of explanatory variables. This paper proposes a straightforward means of postprocessing posterior estimates of a conjugate Bayesian VAR to effectively perform equation‐specific covariate selection. Compared with existing techniques using shrinkage alone, our approach combines shrinkage and sparsity in both the VAR coefficients and the error variance–covariance matrices, greatly reducing estimation uncertainty in large dimensions while maintaining computational tractability. We illustrate our approach by means of two applications. The first application uses synthetic data to investigate the properties of the model across different data‐generating processes, and the second application analyzes the predictive gains from sparsification in a forecasting exercise for U.S. data.

show abstract

Time‐varying partial‐directed coherence approach to forecast global energy prices with stochastic volatility model

Dhifaoui,

Ben Jabeur,

Khalfaoui

et al. 2023

Journal of Forecasting

View full text Add to dashboard Cite

For investors and policymakers, forecasting energy prices with accuracy is essential and plays a major role in the global bulk commodity markets. The current study proposes a novel hybrid forecasting model to predict global energy prices, namely, time‐varying partial‐directed coherence with stochastic volatility. The proposed method combines partial‐directed coherence analysis and stochastic volatility models. Accordingly, this study attempts to provide an in‐depth understanding of the relationship between energy markets and global economic conditions as well as the causality pathway between the underlined markets. Monthly data from January 1982 to July 2022 is used in this study. The results show a strong causality between global economic conditions, European oil, and natural gas prices and have profound implications for policymakers. For completeness, we extend the analysis to the forecasting ability of global economic conditions for oil and natural gas prices. The out‐of‐sample results show that the autoregressive model incorporating the global economic conditions index can significantly improve the accuracy of oil and gas price forecasts. In addition, our results are strongly robust over a variety of time horizons for forecasting, and they provide valuable insights into the forecasting choices to guide investment strategies in the energy and financial market.

show abstract

Inducing Sparsity and Shrinkage in Time-Varying Parameter Models

Cited by 65 publications

References 47 publications

An Overview of Dynamic Model Averaging Techniques in Time‐series Econometrics

An Overview of Dynamic Model Averaging Techniques in Time‐series Econometrics

Combining shrinkage and sparsity in conjugate vector autoregressive models

Time‐varying partial‐directed coherence approach to forecast global energy prices with stochastic volatility model

Contact Info

Product

Resources

About