Residual-based regime detection using local walk-forward autoregressive models; Молодежь и современные информационные технологии

Bibliographic Details
Parent link:	Молодежь и современные информационные технологии.— 2026.— С. 496-500
Main Author:	Khalil M. E. T.
Other Authors:	Aksenov A. V. Andrey Vladimirovich (научный руководитель)
Summary:	Financial markets are non-stationary, causing forecasting models to fail intermittently. This paper proposes a walk-forward ARIMAX framework where residual dynamics, rather than forecasts alone, are used to detect regime changes. Empirical results on hourly XAUUSD data demonstrate that normalized forecast errors provide informative, model-aware regime signals Текстовый файл
Language:	English
Published:	2026
Subjects:	электронный ресурс труды ученых ТПУ regime detection financial markets residual analysis time series финансовые рынки остаточный анализ временные ряды
Online Access:	http://earchive.tpu.ru/handle/11683/139121
Format:	Electronic Book Chapter
KOHA link:	https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=687470

Description
Summary:	Financial markets are non-stationary, causing forecasting models to fail intermittently. This paper proposes a walk-forward ARIMAX framework where residual dynamics, rather than forecasts alone, are used to detect regime changes. Empirical results on hourly XAUUSD data demonstrate that normalized forecast errors provide informative, model-aware regime signals Текстовый файл

Similar Items