2026
Topological and Self-Structured Approaches to Supervised Anomaly Detection in Econometrics
Abstract: This article proposes a robust econometric framework for anomaly detection in nonstationary time series affected by noise, outliers, and regime shifts. The method combines windowed feature construction, supervised learning, and stability-oriented regularization, while enabling optional topological and structural diagnostics to corroborate detected transitions. A reproducible pipeline trains models, calibrates decision thresholds, and preserves artifacts for transparent validation, including metrics, figures, a…
This publication either has no citations yet, or we are still processing them
Set email alert for when this publication receives citations?
