{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T15:58:42Z","timestamp":1792166322525,"version":"4.4.0"},"posted":{"date-parts":[[2026]]},"group-title":"SSRN","reference-count":25,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/www.uspto.gov\/ip-policy\/copyright-policy\/copyright-basics"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>This study examines whether machine-learning models provide additional value beyond the VIX in forecasting next-day extreme market stress in the S&amp;P 500. Using daily data from 2005\u20132025, we define extreme stress based on high volatility conditions and large market declines and compare logistic regression, random forests, gradient-boosted trees, and a class-weighted LSTM with three benchmarks: persistence, the VIX level, and a simple VIX\u2013return model. Models are evaluated using chronological hold-out and expanding-window walk-forward tests with bootstrap inference. The results show that the VIX remains a challenging benchmark, achieving a hold-out ROC-AUC of 0.940 compared with 0.915 for the best machine-learning model and 0.830 for the LSTM. Tree-based models do not outperform the VIX in walk-forward evaluations, especially for newly emerging stress episodes. Adding economic policy uncertainty and geopolitical risk provides only marginal predictive improvements, increasing AUC by no more than 0.005. Robustness tests using a VIX-independent crash-day definition confirm the importance of volatility-based information. These findings suggest that machine-learning models should be assessed against economically meaningful benchmarks, as market-implied volatility already captures substantial information about short-term stress risk.<\/jats:p>","DOI":"10.2139\/ssrn.7620448","type":"posted-content","created":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T15:38:17Z","timestamp":1792165097000},"source":"Crossref","is-referenced-by-count":0,"title":["Do Machine Learning Models Add Value Beyond the VIX in Forecasting Extreme Market Stress?"],"prefix":"10.2139","author":[{"given":"Raymond","family":"Lim","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"ref1","article-title":"RF: adding EPU","journal-title":"Notes: Entries are the difference in AUC between the first and second model"},{"key":"ref2","doi-asserted-by":"crossref","DOI":"10.1016\/j.frl.2026.110322","article-title":"State economic policy uncertainty and municipal bond pricing","volume":"106","author":"A A Abakah","year":"2026","journal-title":"Finance Research Letters"},{"key":"ref3","article-title":"The dynamic relationship among private and public markets and its most important features","volume":"89","author":"A Aslam","year":"2026","journal-title":"International Business and Finance"},{"key":"ref4","doi-asserted-by":"crossref","DOI":"10.1016\/j.mlwa.2026.100840","article-title":"Multimodal information fusion for financial forecasting via cross-attention and calibrated uncertainty","volume":"23","author":"J Bustarviejo","year":"2026","journal-title":"Machine Learning with Applications"},{"key":"ref5","article-title":"SVIX, VIX, and cryptocurrency market return","volume":"104","author":"I.-C Chiu","year":"2025","journal-title":"The Quarterly Review of Economics and Finance"},{"key":"ref6","article-title":"Geopolitical risk and energy market tail risk forecasting: An explainable machine learning approach","volume":"39","author":"M A F Chowdhury","year":"2025","journal-title":"Journal of Commodity Markets"},{"key":"ref7","doi-asserted-by":"crossref","DOI":"10.1016\/j.frl.2025.108508","article-title":"Machine learning models to predict stock market spillovers","volume":"86","author":"L Del Nero","year":"2025","journal-title":"Finance Research Letters"},{"key":"ref8","doi-asserted-by":"crossref","DOI":"10.1016\/j.dajour.2026.100742","article-title":"A machine learning framework for short-horizon monitoring of systemic financial stress in Europe","volume":"20","author":"D Diachkov","year":"2026","journal-title":"Decision Analytics Journal"},{"key":"ref9","doi-asserted-by":"crossref","DOI":"10.1016\/j.intfin.2024.101975","article-title":"Forecasting international financial stress: The role of climate risks","volume":"92","author":"S Fava","year":"2024","journal-title":"Journal of International Financial Markets, Institutions and Money"},{"key":"ref10","doi-asserted-by":"crossref","DOI":"10.1016\/j.jimonfin.2025.103467","article-title":"Your fear is (partly) mine: the role of non-VIX volatility in forecasting regional stock market volatility using interpretable machine learning","volume":"160","author":"L Feng","year":"2026","journal-title":"Journal of International Money and Finance"},{"key":"ref11","article-title":"Beyond conventional sentiment indicators: Cryptocurrency's hidden potential in VIX forecasting","volume":"161","author":"M Gu","year":"2026","journal-title":"Economic Modelling"},{"key":"ref12","doi-asserted-by":"crossref","DOI":"10.1016\/j.frl.2026.110096","article-title":"Tail risk forecasting for crude oil futures under market stress: New insights from MES regression","volume":"102","author":"S Hong","year":"2026","journal-title":"Finance Research Letters"},{"key":"ref13","article-title":"Geopolitical risk and economic policy uncertainty aggravate the vulnerability of crude oil volatility","volume":"110","author":"G Hu","year":"2026","journal-title":"The Quarterly Review of Economics and Finance"},{"key":"ref14","article-title":"Learn to explain the smile: An interpretable hybrid machine learning model to understand the implied volatility of CSI 300 options","volume":"96","author":"P Li","year":"2026","journal-title":"Pacific-Basin Finance Journal"},{"key":"ref15","article-title":"News-based equity market uncertainty aligned: An informative predictor for gold market volatility","volume":"40","author":"S Li","year":"2025","journal-title":"Journal of Commodity Markets"},{"key":"ref16","doi-asserted-by":"crossref","DOI":"10.1016\/j.eneco.2024.107980","article-title":"What drives the uranium sector risk? The role of attention, economic and geopolitical uncertainty","volume":"140","author":"\u0160 Ly\u00f3csa","year":"2024","journal-title":"Energy Economics"},{"key":"ref17","doi-asserted-by":"crossref","DOI":"10.1016\/j.eswa.2023.122896","article-title":"A hybrid model integrating artificial neural network with multiple GARCH-type models and EWMA for performing the optimal volatility forecasting of market risk factors","volume":"243","author":"F P\u00e9rez-Hern\u00e1ndez","year":"2024","journal-title":"Expert Systems with Applications"},{"key":"ref18","author":"M Shafiei","year":"2026","journal-title":"Uncovering heterogeneous effects of global uncertainty on stock returns using causal forest double machine learning. Expert Systems with Applications"},{"key":"ref19","doi-asserted-by":"crossref","DOI":"10.1016\/j.irfa.2025.104457","article-title":"Semiconductor game of thrones: A comprehensive study of geopolitical and equity market uncertainty transmission","volume":"106","author":"M Snarska","year":"2025","journal-title":"International Review of Financial Analysis"},{"key":"ref20","doi-asserted-by":"crossref","DOI":"10.1016\/j.ribaf.2024.102448","article-title":"Recession fears and stock markets: An application of directional wavelet coherence and a machine learning-based economic agent-determined Google fear index","volume":"72","author":"J J Szczygielski","year":"2024","journal-title":"Research in International Business and Finance"},{"key":"ref21","article-title":"Economic policy uncertainty and its effect on stocks, bitcoin and the gold markets","author":"K Upadhyaya","year":"2026","journal-title":"Journal of Financial Economic Policy"},{"key":"ref22","article-title":"Stock return forecasting under climate policy uncertainty: A regime-dependent model comparison","volume":"108","author":"A S Wafi","year":"2026","journal-title":"Finance Research Letters"},{"key":"ref23","doi-asserted-by":"crossref","DOI":"10.1016\/j.qref.2025.102026","article-title":"Predicting abnormal capital flow episodes with machine learning methods","volume":"103","author":"B Wang","year":"2025","journal-title":"The Quarterly Review of Economics and Finance"},{"key":"ref24","doi-asserted-by":"crossref","DOI":"10.1016\/j.jbankfin.2026.107719","article-title":"Joint valuation of SPX and VIX options by GARCH models with bad and good environments","volume":"188","author":"Z Wang","year":"2026","journal-title":"Journal of Banking & Finance"},{"key":"ref25","doi-asserted-by":"crossref","DOI":"10.1016\/j.najef.2025.102554","article-title":"Time-frequency quantile effect of global uncertainty on stock markets: evidence from wavelet decomposition","volume":"81","author":"J Yuan","year":"2026","journal-title":"The North American Journal of Economics and Finance"}],"container-title":[],"original-title":[],"deposited":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T15:40:43Z","timestamp":1792165243000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ssrn.com\/abstract=7620448"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":25,"URL":"https:\/\/doi.org\/10.2139\/ssrn.7620448","relation":{},"subject":[],"published":{"date-parts":[[2026]]},"subtype":"preprint"}}