Choosing Prior Hyperparameters
With Applications to Time-Varying Parameter Models
Bibliographic Data
| ID | 19417520 |
|---|---|
| Authors | Pooyan Amir-Ahmadi (Department of Economics, University of Illinois at Urbana-Champaign, Urbana, IL 61820 ()), Christian Matthes (Federal Reserve Bank of Richmond, Richmond, VA 23219 ()), Mu-Chun Wang (Universität Hamburg), Mu‐Chun Wang (0000-0002-4605-0658, Universität Hamburg) |
| Year | 2020 |
| Volume | 38 |
| Issue | 1 |
| Pages | 124-136 |
| Publication date | 2020-01-02 |
| Open Access | No |
| Type | ARTICLE |
| Venue | Journal of Business and Economic Statistics (JOURNAL) |
| Journal identifiers | ISSN: 0735-0015 • E-ISSN: 1537-2707 |
| Publisher | Informa UK Limited (PUBLISHER • GB) |
| DOI | 10.1080/07350015.2018.1459302 |
| OpenAlex | W2803100032 |
| Language | EN |
| Citations received | 8 |
| References cited | 18 |
Time-varying parameter models with stochastic volatility are widely used to study macroeconomic and financial data. These models are almost exclusively estimated using Bayesian methods. A common practice is to focus on prior distributions that themselves depend on relatively few hyperparameters such as the scaling factor for the prior covariance matrix of the residuals governing time variation in the parameters. The choice of these hyperparameters is crucial because their influence is sizeable for standard sample sizes. In this article, we treat the hyperparameters as part of a hierarchical model and propose a fast, tractable, easy-to-implement, and fully Bayesian approach to estimate those hyperparameters jointly with all other parameters in the model. We show via Monte Carlo simulations that, in this class of models, our approach can drastically improve on using fixed hyperparameters previously proposed in the literature.
Bayesian inference · Bayesian probability · Covariance · Econometrics · Hyperparameter · Machine learning · Monte Carlo method · Statistics · Stochastic volatility · Computer Science · Financial Risk and Volatility Modeling · Forecasting Techniques and Applications · Mathematics · Monetary Policy and Economic Impact · Artificial Intelligence
Monetary Policy Uncertainty and Inflation Expectations
The macroeconomy as a random forest
Identifying exchange rate effects and spillovers of US monetary policy shocks in the presence of time‐varying instrument relevance
High‐Frequency Instruments With Time‐Varying Reliability
The Macroeconomic Effects of International Food Price Shocks on China
On the Time‐Varying Effects of Economic Policy Uncertainty on the US Economy
Fast, Order-Invariant Bayesian Inference in VARs Using the Eigendecomposition of the Error Covariance Matrix
Estimating Posterior Sensitivities with Application to Structural Analysis of Bayesian Vector Autoregressions
Large Bayesian vector auto regressions
Drifts and volatilities
Time Varying Structural Vector Autoregressions and Monetary Policy
Forecasting and conditional projection using realistic prior distributions
Evidence on Structural Instability in Macroeconomic Time Series Relations
A Century of Inflation Forecasts
Prior Selection for Vector Autoregressions
| Citations per year | 1.33 |
|---|---|
| Citation span | 2020 - 2026 (7) |
| Highly cited | No |