Fast and Flexible Bayesian Inference in Time-varying Parameter Regression Models

  title={Fast and Flexible Bayesian Inference in Time-varying Parameter Regression Models},
  author={Niko Hauzenberger and Florian Huber and Gary Koop and Luca Onorante},
  journal={Journal of Business \& Economic Statistics},
In this paper, we write the time-varying parameter regression model involving K explanatory variables and T observations as a constant coefficient regression model with TK explanatory variables. In contrast with much of the existing literature which assumes coefficients to evolve according to a random walk, this specification does not restrict the form that the time-variation in coefficients can take. We develop computationally efficient Bayesian econometric methods based on the singular value… 

Figures and Tables from this paper

Bayesian Inference in High-Dimensional Time-varying Parameter Models using Integrated Rotated Gaussian Approximations
This paper develops computationally efficient Bayesian methods for estimating TVP models using an integrated rotated Gaussian approximation (IRGA), exploiting the fact that whereas constant coefficients on regressors are often important, most of the TVPs are often unimportant.
General Bayesian time-varying parameter VARs for predicting government bond yields
Time-varying parameter (TVP) regressions commonly assume that timevariation in the coefficients is determined by a simple stochastic process such as a random walk. While such models are capable of
Dynamic Shrinkage Priors for Large Time-varying Parameter Regressions using Scalable Markov Chain Monte Carlo Methods
A new dynamic shrinkage prior is proposed which reflects the empirical regularity that TVPs are typically sparse and is found to effectively pick out small amounts of parameter change and the methods to forecast well.
Sparse time-varying parameter VECMs with an application to modeling electricity prices
This paper combines a set of econometric techniques for dynamic model specification in an automatic fashion and employs continuous global-local shrinkage priors for pushing the parameter space towards sparsity, alleviating overfitting concerns and reduces parameter estimation uncertainty.
Flexible Mixture Priors for Time-varying Parameter Models
Time-varying parameter (TVP) models often assume that the TVPs evolve according to a random walk. This assumption, however, might be questionable since it implies that coefficients change smoothly
Real-time inflation forecasting using non-linear dimension reduction techniques
Forecasting euro area inflation using a huge panel of survey expectations
. In this paper, we forecast euro area inflation and its main components using an econometric model which exploits a massive number of time series on survey expectations for the European Commission’s
Modeling tail risks of inflation using unobserved component quantile regressions


Forecasting and Conditional Projection Using Realistic Prior Distributions
A forecasting procedure based on a Bayesian method for estimating vector autoregressions is applied to ten macroeconomic variables and is shown to improve out-of-sample forecasts relative to univariate equations.
Inducing Sparsity and Shrinkage in Time-Varying Parameter Models
Abstract Time-varying parameter (TVP) models have the potential to be over-parameterized, particularly when the number of variables in the model is large. Global-local priors are increasingly used to
Shrinkage in the Time-Varying Parameter Model Framework Using the R Package shrinkTVP
The R package shrinkTVP is introduced, which provides a fully Bayesian implementation of shrinkage priors for TVP models, taking advantage of recent developments in the literature, in particular that of Bitto and Fr\"uhwirth-Schnatter (2019).
Forecasting Inflation in a Data-Rich Environment: The Benefits of Machine Learning Methods
It is shown that ML models with a large number of covariates are systematically more accurate than the benchmarks and the ML method that deserves more attention is the random forest model, which dominates all other models.
Phillips Curves in the Euro Area
We perform a robust estimation of the Phillips curve in the euro area using a battery of 630 theory-driven models. We extend the existing literature by adding model specifications, taking into
LR-GLM: High-Dimensional Bayesian Inference Using Low-Rank Data Approximations
LR-GLM provides a full Bayesian posterior approximation and admits running times reduced by a full factor of the parameter dimension and shows how the choice of rank allows a tunable computational-statistical trade-off.
On Gibbs sampling for state space models
SUMMARY We show how to use the Gibbs sampler to carry out Bayesian inference on a linear state space model with errors that are a mixture of normals and coefficients that can switch over time. Our
Online Prediction Under Model Uncertainty via Dynamic Model Averaging: Application to a Cold Rolling Mill
DMA over a large model space led to better predictions than the single best performing physically motivated model, and it recovered both constant and time-varying regression parameters and model specifications quite well.
Inference with normal-gamma prior distributions in regression problems
This paper considers the efiects of placing an absolutely continuous prior distribution on the regression coe-cients of a linear model. We show that the posterior expectation is a matrix-shrunken