Time-series modeling plays a crucial role in analyzing and forecasting financial volatility. Classical approaches, such as the Autoregressive (AR) and Fractional Autoregressive (FAR) models, capture short-term linear dependencies and long-range correlations, respectively, but their reliance on fixed structures and stationarity assumptions limits their adaptability to evolving market dynamics. To overcome these limitations, this study introduces a Fractional Time-Varying Autoregressive (FTVAR) framework that allows model parameters to evolve smoothly over time, integrating long-memory effects with nonstationary behavior. The FTVAR process is examined through two complementary methods: a Generalized Additive Model (GAM) for interpretable estimation of time-varying coefficients, and a Physics-Informed Neural Network (PINN) that embeds system dynamics to enhance forecasting under complex conditions. Simulation studies demonstrate that the FTVAR model consistently outperforms conventional AR approaches, offering superior forecasting accuracy, robustness to nonstationarity, and a more comprehensive representation of evolving volatility structures. Empirical analyses on the S&P 500 and VIX indices further confirm the effectiveness and practical relevance of the proposed framework.
We conduct cluster analysis of a class of locally asymptotically self-similar stochastic processes with finite covariance structures, which includes Brownian motion, fractional Brownian motion, and multifractional Brownian motion as paradigmatic examples. Given the true number of clusters, a new covariance-based dissimilarity measure is introduced, based on which we obtain approximately asymptotically consistent algorithms for clustering locally asymptotically self-similar stochastic processes. In the simulation study, clustering data sampled from fractional and multifractional Brownian motions with distinct Hurst parameters illustrates the approximated asymptotic consistency of the proposed algorithms. Clustering global financial markets’ equity indexes returns and sovereign CDS spreads provides a successful real world application. Implementations in MATLAB of the proposed algorithms and the simulation study are publicly shared in GitHub.
We introduce a new unsupervised learning problem: clustering wide-sense stationary ergodic stochastic processes. A covariance-based dissimilarity measure together with asymptotically consistent algorithms is designed for clustering offline and online datasets, respectively. We also suggest a formal criterion on the efficiency of dissimilarity measures, and discuss of some approach to improve the efficiency of our clustering algorithms, when they are applied to cluster particular type of processes, such as self-similar processes with wide-sense stationary ergodic increments. Clustering synthetic data and real-world data are provided as examples of applications.
Detrended fluctuation analysis (DFA) and its multifractal extension (MF-DFA) estimate scaling exponents from a fixed segmentation of the data. On a nonstationary series, however, a segment that crosses a transition of the local Hurst function mixes two scaling regimes and biases the estimate. We propose a local Hurst-guided segmentation for MF-DFA that leaves the classical fluctuation function and polynomial detrending unchanged and adapts only the grouping of the observations to : a rolling DFA curve is estimated, its transitions are located by CUSUM or PELT, and the multifractal spectrum is estimated separately within each detected regime, with a boundary inset that discards observations of uncertain regime membership. When is constant the procedure reduces to standard MF-DFA. We prove that regime mixing induces a spurious multifractal width in global MF-DFA, and we bound the error that a misplaced transition propagates into the state-level and show that a boundary inset of the order of the detection error removes it. Simulations use multifractional Brownian motion with stepwise and continuously varying Hurst functions. The segmented estimators recover the state spectra and remove the spurious multifractal width that global MF-DFA attributes to the series as a whole. An application to the daily S&P 500 index over 1999–2026 completes the study.