What question did this study set out to answer?

This work aims to quantify the effect of USA tariff announcements on the S&P 500 and S&P/ASX 200 stock indices using advanced statistical methods.

February 2, 2026Open Access

Quantifying USA tariffs effect: machine learning, entropy and fractal insights into the stock markets

Key Points

This work aims to quantify the effect of USA tariff announcements on the S&P 500 and S&P/ASX 200 stock indices using advanced statistical methods.
Applied normalized permutation entropy to evaluate ordinal-pattern complexity.
Used fractal dimension estimates from Higuchi, Katz, and Sevcik methods for geometric scaling analysis.
Employed machine learning models to predict market responses under various conditions.
Analyzed post-event market dynamics across different time-windows.
Post-event permutation entropy values remained high, indicating stable distributional entropy.
Higuchi fractal dimension increased, suggesting minor changes in high-frequency market behavior.
Katz and Sevcik dimensions showed a decrease, indicating a smoother medium-scale structure.
High cross-market predictability was achieved with ensemble machine learning models, with $R^2>0.98$.

Abstract

Abstract This study presents a multiscale econometric framework to evaluate the impact of the USA tariff announcement of 2 April 2025 on the S AUS: 1. 524 1. 533), indicating a marginal rise in high-frequency roughness, while Katz and Sevcik dimensions decline (USA Sevcik: 1. 337 1. 229), consistent with smoother medium-scale structure. Across all measures, the shock does not generate a statistically meaningful structural break. Additionally, machine-learning models (kNN, SVR, Random Forest, XGBoost, Neural Network) demonstrate strong cross-market predictability, with ensemble methods achieving out-of-sample R²>0. 98. Overall, the results suggest that both markets absorbed the tariff shock rapidly, exhibiting stable multiscale dynamics despite heightened geopolitical uncertainty. A model-agnostic XAI framework combining SHAP attribution and permutation-based diagnostics is used to isolate robust and independent information content.

Read Full Paperexternally

Bookmark

View Full Paper

Cite This Study

Mridul Patel (Thu,) studied this question.

synapsesocial.com/papers/6980fc17c1c9540dea80dd94 https://doi.org/https://doi.org/10.1007/s11135-026-02581-9

Bookmark

View Full Paper