Algorithmic Monitoring: Measuring Market Stress with Machine Learning
Marc Schmitt \ORCID 0000-0003-4550-2963 · 2026
Read Algorithmic Monitoring: Measuring Market Stress with Machine Learning by Marc Schmitt \ORCID 0000-0003-4550-2963 (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-07066.txt.
Description
Marc Schmitt builds an index that gauges the odds of high stress in the United States equity market one month ahead. Each month, he sums up the cross-section of single stocks into a set of interpretable fragility signals, drawn from CRSP daily data. Those signals feed a logistic regression with an L1 penalty. The model is refit as months roll forward in real time, yielding a forward-looking stress probability.
Out of sample, the index tracks major stress episodes. It beats a lean benchmark built on lagged market return and realized volatility, on both discrimination and accuracy. It gives calibrated probabilities on an economically meaningful scale. Schmitt also shows how the index can serve as a probability tool for measuring, for use in financial econometrics research. The result is a transparent gauge of near-term equity-market stress risk that is easy to update.
Catalog record
| CREATOR | Marc Schmitt \ORCID 0000-0003-4550-2963 |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-07066