Skip to content
Preprint

Scale or speed? When do path signatures improve volatility-regime forecasts?

Oct 2026 · 0 citations · 24 references
Mathematics

Abstract

We study when path signatures, increasingly used as model-free features of return paths, improve out-of-sample forecasts of volatility regimes. In daily United States and developed-market equity returns from 2005 to 2025, machine-learning classifiers do not beat a heterogeneous autoregressive (HAR) regression of realized volatility, although signatures improve the classifiers'accuracy and probability forecasts in the United States. Adding raw signatures to the HAR regression itself improves United States forecasts of next-month volatility terciles, raising balanced accuracy by 2.7 percentage points and lowering the ranked probability score after multiple-testing correction, whereas adding conventional volatility features does not; outside the United States no gain survives correction. Controlled Heston experiments explain the pattern: signatures add modestly when regimes differ in variance level, which realized volatility already captures, and more when regimes share one stationary variance law but differ in mean-reversion speed. Better regime identification does not, however, deliver earlier regime-change signals, and no accuracy gain translates into a resolved Sharpe-ratio gain in a volatility-managed portfolio. Signatures thus add information about how volatility evolves, which realized-volatility predictors miss.

View source

Similar papers

Open access Oct 2026

Trend-Following Indicators Under Stable and High-Volatility Market Regimes: An Empirical Assessment Using GARCH-Based Volatility Classification

This study investigates whether the directional accuracy of commonly used technical indicators changes with market volatility and whether combining several indicators can improve signal reliability. The analysis draws on 500 daily financial observations and considers moving averages, the Relative Strength Index (RSI),...

Rania Loubaris, Almahdi Koraich · 0 citations
Preprint Oct 2026

Flowz: When Does Learning Forecast Errors Add Value? Conditional Flow Matching for Financial Price Forecasts

Learning a forecasting model's errors offers a modular way to improve its predictive distribution while retaining its underlying dynamics. We study this approach with Flowz, a conditional flow model of direct multi-step price-level residuals fitted using random features and ridge regression. Each EWMA, ARIMA, ARFIMA, a...

G. Vercellino · 0 citations
Open access Oct 2026

Bells and Whistles of Nowcasting Models: Which Matter and When?

This paper investigates which features of a Bayesian dynamic factor model improve U.S. GDP nowcasts and whether their contribution varies across historical episodes. Using pseudo-real-time information sets for 25 series, we estimate 16 combinations of dynamic heterogeneity, stochastic volatility, time-varying long-run...

Freddy García-Albán, Manuel González-Astudillo · 0 citations
Preprint Oct 2026

Volatility-Corrected Prediction Intervals: Locally Adaptive Uncertainty Quantification for Probabilistic Time-Series Forecasting

Prediction intervals from probabilistic time-series forecasters are commonly assessed by marginal coverage, the long-run proportion of observations falling inside the intervals. However, forecast errors in time series are often serially dependent and heteroskedastic, so prediction intervals can attain near-nominal marg...

Jason B. Cho, David S. Matteson · 0 citations
Conference Open access 2026

Factor-Attenuation-Driven Dispersion in Trading-Factor Returns: Evidence From Volatility-Dependent Market States

With the development of quantitative investment and asset pricing, trading factors are widely used in explaining cross-sectional stock returns and constructing portfolios. This paper constructs monthly samples from daily data of US stocks over 2000–2025 and selects five trading factors, namely price delay, low idiosync...

Tai-Heng Cao · 0 citations

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