Philippe Goulet Coulombe, Maxime Leroux, Dalibor Stevanovic, Stéphane Surprenant
arXiv 28 Aug 2020 · Econometrics · publishedJournal of Applied Econometrics (2022) · 222 citations (OpenAlex)
arXiv:2008.12477 · PDF · DOI · OpenAlex · Extracted main text
We move beyond "Is Machine Learning Useful for Macroeconomic Forecasting?" by adding the "how". The current forecasting literature has focused on matching specific variables and horizons with a particularly successful algorithm. In contrast, we study the usefulness of the underlying features driving ML gains over standard macroeconometric methods. We distinguish four so-called features (nonlinearities, regularization, cross-validation and alternative loss function) and study their behavior in both the data-rich and data-poor environments. To do so, we design experiments that allow to identify the "treatment" effects of interest. We conclude that (i) nonlinearity is the true game changer for macroeconomic prediction, (ii) the standard factor model remains the best regularization, (iii) K-fold cross-validation is the best practice and (iv) the $L_2$ is preferred to the $\bar \epsilon$-insensitive in-sample loss. The forecasting gains of nonlinear techniques are associated with high macroeconomic uncertainty, financial stress and housing bubble bursts. This suggests that Machine Learning is useful for macroeconomic forecasting by mostly capturing important nonlinearities that arise in the context of uncertainty and financial frictions.
appendix boundary found by appendix_command · 51% 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 | Sermpinis, G., Stasinakis, C., Theofilatos, K., and Karathanasopoulo… (2014) Inflation and unemployment forecasting with genetic support vector regression | 1.000 | 5 | 3 | 100% |
| 2 | Stock, J. H. and Watson, M. W (2002) Macroeconomic forecasting using diffusion indexes | 0.928 | 5 | 4 | 80% |
| 3 | Medeiros, M. C., Vasconcelos, G. F., Veiga, Á., and Zilberman, E (2019) Forecasting Inflation in a Data-Rich Environment: The Benefits of Machine Learning Methods | 0.928 | 4 | 3 | 100% |
| 4 | Gu, S., Kelly, B., and Xiu, D (2020) Empirical Asset Pricing via Machine Learning | 0.843 | 4 | 3 | 75% |
| 5 | Stock, J. H. and Watson, M. W (2012) Generalized Shrinkage Methods for Forecasting Using Many Predictors | 0.843 | 3 | 3 | 100% |
| 6 | Hastie, T., Tibshirani, R., and Friedman, J (2009) The Elements of Statistical Learning: Data Mining, Interference, and Prediction | 0.811 | 4 | 2 | 100% |
| 7 | McCracken, M. W. and Ng, S (2016) FRED-MD: A Monthly Database for Macroeconomic Research | 0.737 | 3 | 3 | 67% |
| 8 | Diebold, F. X. and Mariano, R. S (1995) Comparing predictive accuracy | 0.737 | 3 | 2 | 100% |
| 9 | Gu, S., Kelly, B., and Xiu, D (2020) Autoencoder asset pricing models | 0.737 | 3 | 2 | 100% |
| 10 | Marcellino, M (2008) A linear benchmark for forecasting GDP growth and inflation? | 0.737 | 3 | 2 | 100% |
Showing the top 10 of 98 scored citations.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.
| Citing paper | Intensity | Mentions | Sections | |
|---|---|---|---|---|
| 1 | 2512.02092 | 1.000 | 8 | 5 |
| 2 | Macroeconomic Data Transformations Matter | 1.000 | 7 | 4 |
| 3 | Can Machine Learning Catch the COVID-19 Recession? | 1.000 | 5 | 3 |
| 4 | Nowcasting Italian Municipal Income with Nightlights: A Deep Learning Approach | 0.928 | 5 | 5 |
| 5 | Explainable Machine Learning for Macroeconomic and Financial Nowcasting | 0.874 | 5 | 2 |
| 6 | The Macroeconomy as a Random Forest | 0.843 | 5 | 3 |
| 7 | 0cm To Bag is to Prune | 0.737 | 3 | 2 |
| 8 | 0.54cm Time-Varying Parameters as Ridge Regressions | 0.737 | 3 | 2 |
| 9 | The Kernel Trick for Nonlinear Factor Modeling | 0.644 | 2 | 2 |
| 10 | 0cm Slow-Growing Trees | 0.644 | 2 | 2 |