FINANCIAL STRESS AND COMMODITY DYNAMIC: A WAVELET-BASED ANOMALY DETECTION APPROACH

Abstract

This study investigates the dynamic interdependence between the Indian Financial Stress Index (IFSI) and major global commodity prices—gold, silver, copper, crude oil, and natural gas—using a wavelet-based analytical framework. Given the inherent volatility and non-stationarity of financial and commodity markets, wavelet analysis provides an effective approach to capture their evolving, multi-scale interactions. The research employs the Daubechies wavelet Discrete Wavelet Transform (DWT) to upscale monthly data into weekly series through interpolation, thereby enhancing the granularity of insights. Continuous Wavelet Transform (CWT) and Wavelet Coherence (WTC) with the Morlet wavelet are applied for time–frequency decomposition and phase difference analysis, enabling detection of bi-directional lead–lag relationships between IFSI and commodity prices. Results highlight that financial stress indicators alternately lead or lag commodity price movements, particularly during high-volatility episodes such as the Global Financial Crisis (2008–09) and the COVID-19 pandemic (2020). Monte Carlo simulations confirm the statistical significance of these dynamics. Furthermore, wavelet-based anomaly detection successfully identifies abnormal fluctuations in commodity prices aligned with stress-induced market shifts, providing potential early-warning signals of distress. The findings offer valuable implications for risk management, investment strategies, and policy interventions in emerging economies like India.