Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Malcom: Indonesian Journal of Machine Learning and Computer Science
Firdaus, Nelwan Topan
Unknown Affiliation
Computer Science & IT

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Early Warning Systems for Financial Crisis Prediction: A Systematic Literature Review of Econometrics, Machine Learning and Uncertainty Indices Firdaus, Nelwan Topan; Santoso, Noviyanti
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 5 No. 4 (2025): MALCOM October 2025
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v5i4.2314

Abstract

This study evaluates the integration of econometric methods, machine learning models, and uncertainty indices within the framework of Early Warning Systems (EWS) for financial crisis prediction in stock markets. A Systematic Literature Review (SLR) was conducted on studies published between 2008 and 2024, sourced from reputable databases such as Scopus, IEEE, and other international publishers. The review identifies three main objectives. First, the development of predictive models for market volatility and systemic risk using econometric and machine learning approaches. Second, the diagnosis of risk factors by incorporating macroeconomic indicators, commodity prices, geopolitical uncertainty, and sentiment data from big data sources. Third, the evaluation of policy implications and the role of composite indicators in maintaining financial stability. The dominant data categories include market data (prices, returns, volatility, sectoral indices), macroeconomic indicators (production, interest rates, leading indicators), commodities and energy (oil and gold), and measures of risk and uncertainty (GPR, GEPU, TPU, sentiment). Methodologically, studies employ time series econometrics (ARIMA, GARCH, VAR, spillover), machine learning, hybrid approaches, and indicator or policy-based frameworks. The findings reveal a growing trend toward multivariate and hybrid models, yet highlight limited integration across methods and data domains. 
Co-Authors Santoso, Noviyanti