Niko Hauzenberger, Florian Huber, Karin Klieber, Massimiliano Marcellino
arXiv 9 Nov 2022 · Econometrics · publishedJournal of Econometrics (2024) · 17 citations (OpenAlex)
arXiv:2211.04752 · PDF · DOI · OpenAlex · Extracted main text
Macroeconomic data is characterized by a limited number of observations (small T), many time series (big K) but also by featuring temporal dependence. Neural networks, by contrast, are designed for datasets with millions of observations and covariates. In this paper, we develop Bayesian neural networks (BNNs) that are well-suited for handling datasets commonly used for macroeconomic analysis in policy institutions. Our approach avoids extensive specification searches through a novel mixture specification for the activation function that appropriately selects the form of nonlinearities. Shrinkage priors are used to prune the network and force irrelevant neurons to zero. To cope with heteroskedasticity, the BNN is augmented with a stochastic volatility model for the error term. We illustrate how the model can be used in a policy institution by first showing that our different BNNs produce precise density forecasts, typically better than those from other machine learning methods. Finally, we showcase how our model can be used to recover nonlinearities in the reaction of macroeconomic aggregates to financial shocks.
appendix boundary found by appendix_command · 84% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | McCracken and Ng (2016) FRED-MD: A monthly database for macroeconomic research | 0.928 | 4 | 3 | 100% |
| 2 | Benigno and Eggertsson (2023) It’s baaack: The surge in inflation in the 2020s and the return of the non-linear Phillips curve | 0.874 | 7 | 2 | 100% |
| 3 | Farrell et al (2021) Deep neural networks for estimation and inference | 0.811 | 4 | 2 | 100% |
| gilchrist2012credit | unmatched citation key gilchrist2012credit | 0.811 | 4 | 2 | 100% |
| polson2018posterior | unmatched citation key polson2018posterior | 0.811 | 4 | 2 | 100% |
| 6 | Chipman et al (2010) BART: Bayesian additive regression trees | 0.737 | 5 | 3 | 40% |
| 7 | Barnichon et al (2022) Are the effects of financial market disruptions big or small? | 0.737 | 3 | 2 | 100% |
| 8 | Bhadra et al (2020) Horseshoe regularisation for machine learning in complex and deep models | 0.644 | 2 | 2 | 100% |
| 9 | Clark et al (2024) Investigating Growth-at-Risk Using a Multicountry Non-parametric Quantile Factor Model | 0.644 | 2 | 2 | 100% |
| 10 | Ghosh et al (2019) Model selection in Bayesian neural networks via horseshoe priors | 0.644 | 2 | 2 | 100% |
Showing the top 10 of 90 scored citations. 2 of these could not be matched to a bibliography entry, so only the citation key is shown.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.
| Citing paper | Intensity | Mentions | Sections | |
|---|---|---|---|---|
| 1 | The ARR2 prior: flexible predictive prior definition for Bayesian auto-regressions | 0.585 | 3 | 1 |
| 2 | Macroeconomic Forecasting and Machine Learning | 0.405 | 1 | 1 |