Skip to content
Log in
Sponsored Artificial Intelligence: A Modern Approach, Global Edition Artificial Intelligence: A Modern Approach, Global Edition

Author

C. Sigauke

View on Semantic Scholar

2 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Open access Jul 2026

Exploring the Dynamics of ZAR/USD Exchange RateVolatility Using the fGARCH and First-Order Beta-Skew-T-EGARCH Models

This study investigates and explores the volatility dynamics of the South African rand against the US dollar (ZAR/USD) using the Family GARCH (fGARCH) model and the First-Order Beta-Skew-T-Generalised Autoregressive Conditional Heteroskedasticity (Beta-Skew-T-EGARCH) model. Currency volatility across the globe, uncertainties, and instability in emerging markets have become increasingly consequential for trade flows, investment allocation, and macroeconomic management. The ZAR/USD serves as a benchmark of South Africa’s economic wealth and vulnerability to external shocks and is one of the most valued, significant, and heavily traded pairings of emerging market currencies. Simple standard GARCH (sGARCH) is one of the most useful models for exchange rate volatility; however, the sGARCH model has some limitations: it fails to accommodate or allow the long memory effects, skewness distribution, and leverage dynamics consistently observed in emerging-market currency returns. This study addresses these limitations by using the fGARCH model, which includes the most popular GARCH models and Beta-Skew-T-EGARCH for daily ZAR/USD returns ranging from 5 January 2000 to 1 October 2024. Five innovation distributions are used for evaluation and comparison under fGARCH and sGARCH, namely generalised hyperbolic (GH), generalised error (GED), skewed Student’s t (SSTD), skewed generalised error (SGED), and Student’s t (STD), with model fitness criteria assessed using the Shibata criterion (SIC), Hannan–Quinn criterion (HQ), Bayesian information criterion (BIC), and Akaike information criterion (AIC), choosing the specification with the lowest overall penalty. It is found that the fGARCH(1,1) model fitted to return-frequency data under the SSTD achieves the lowest AIC, outperforming sGARCH. The study also includes an analysis among covariates, which are day, month, trend, oil, and platinum; the trend variable is a statistically significant predictor, with p = 0.007, showing a positive influence on ZAR/USD volatility. The Beta-Skew-T-EGARCH model with two components divides volatility into long-run and short-run components, which is found to deliver a superior fit over the one-component variant, evidenced by a lower BIC (3.068435) and a higher log-likelihood (−748.464826). The two components confirm that the model captures declining conditional volatility, whereas the one-component model sustains persistence in the evaluated estimates.

Dzulani Mashavhela, Thakhani Ravele, C. Sigauke · 0 citations
Open access Aug 2026

Machine Learning-Based Crisis Detection Framework for Banking Systems: A Case Study of Nigeria

Banking crises are a persistent threat to macroeconomic stability in emerging markets, where conventional econometric monitoring frameworks often fail to capture non-linear macro-financial relationships. This paper examines whether machine learning algorithms can improve the detection of banking crisis risk in Nigeria compared to standard logistic regression. We compare the performance of Random Forest, Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost) against logistic regression using annual data from the African Financial Crises dataset (1954–2014). Resampling is only implemented on the training set to overcome the infrequency of crisis events. Performance on models is assessed based on accuracy, precision, recall, F1-score, and the area under the receiver operating characteristic curve (AUC) in a rigorous out-of-time validation setting. Our findings indicate that tree-based ensemble models outperform logistic regression on the test set: XGBoost achieves the best generalization performance (AUC = 1.0; F1 = 0.95 in non-crisis, 0.80 in crisis), whereas Random Forest has the highest cross-validated F1-score on the training set. The most important variables are exchange rate volatility, inflation, and indicators of systemic crisis. The most significant crisis indicators are, however, seen in crisis years, which means that the annual data do not provide much lead-time to detect the crisis. These results should be taken with caution because of the small sample size and the limited number of crisis observations during the test period. Altogether, machine learning models have potential as additional tools to monitor banking crises in Nigeria, though at the moment they are not fully operational as policy instruments.

Ntanganedzeni Mandiwana, Thakhani Ravele, C. Sigauke et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.