Predicting systemic financial crises with recurrent neural networks

Eero Tölö

Journal of Financial Stability, 2020, vol. 49, issue C

Abstract: We consider predicting systemic financial crises one to five years ahead using recurrent neural networks. We evaluate the prediction performance with the Jórda-Schularick-Taylor dataset, which includes the crisis dates and annual macroeconomic series of 17 countries over the period 1870−2016. Previous literature has found that simple neural net architectures are useful and outperform the traditional logistic regression model in predicting systemic financial crises. We show that such predictions can be significantly improved by making use of the Long-Short Term Memory (RNN-LSTM) and the Gated Recurrent Unit (RNN-GRU) neural nets. Behind the success is the recurrent networks’ ability to make more robust predictions from the time series data. The results remain robust after extensive sensitivity analysis.

Keywords: Early warning system; Systemic Banking crises; Neural networks; Validation (search for similar items in EconPapers)
JEL-codes: C45 C52 G21 (search for similar items in EconPapers)
Date: 2020
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S1572308920300243
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:finsta:v:49:y:2020:i:c:s1572308920300243

DOI: 10.1016/j.jfs.2020.100746

Access Statistics for this article

Journal of Financial Stability is currently edited by I. Hasan, W. C. Hunter and G. G. Kaufman

More articles in Journal of Financial Stability from Elsevier
Bibliographic data for series maintained by Catherine Liu ().