Financial Stability Surveillance Tools: Evaluating the Performance of Stress Indices

Kaelo Ntwaepelo () and Grivas Chiyaba ()
Additional contact information
Kaelo Ntwaepelo: Department of Economics, University of Reading
Grivas Chiyaba: Department of Economics, University of Reading

No em-dp2022-06, Economics Discussion Papers from Department of Economics, University of Reading

Abstract: In this study, we aim to address the emerging debate about whether financial stress indices (FSIs) constructed using advanced methods such as the dynamic factor model and the principal component analysis method, perform better than those aggregated using simple averages, for the case of South Africa. To do so, we construct three FSIs using: the equal-variance weighting method (EVM), the principal component analysis method (PCA) and the dynamic factor model (FAM). We compare the performance of the indices for the period 2009-2020, using four criteria: quantile regressions, ordered probit model, local projections and the autoregressive integrated moving average (ARIMA) forecasting model. The results suggest that FSIs aggregated using the dynamic factor model and the principal component analysis method have a significant comparative advantage in predicting a financial crisis and capturing the vulnerability of the South African financial system to external monetary policy shocks. This suggests that the aggregation method and weighting system involved in constructing a financial stress index affects its performance in monitoring financial stability.

Keywords: financial stress index; forecasts; local projections; ordered probit model; quantile regression (search for similar items in EconPapers)
JEL-codes: B26 C22 C43 C53 E44 E47 (search for similar items in EconPapers)
Pages: 48 pages
Date: 2022-08-24
New Economics Papers: this item is included in nep-ban, nep-cba and nep-fdg
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.reading.ac.uk/web/FILES/economics/emdp202206.pdf

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:rdg:emxxdp:em-dp2022-06

Access Statistics for this paper

More papers in Economics Discussion Papers from Department of Economics, University of Reading Contact information at EDIRC.
Bibliographic data for series maintained by Alexander Mihailov ().