Redefining volatility forecasting in the aerospace and defense sector: application of CEEMDAN-GARCH models

This study pioneers the integration of Complementary Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and advanced GARCH models (IGARCH, SGARCH, and GJR-GARCH) to analyze the volatility of aerospace and defense indices across four countries: China, South Korea, France, and the United Kingdom. Using daily data spanning 2014–2024, the study captures key global disruptions, including the COVID-19 pandemic and the Russia-Ukraine conflict, offering a granular analysis of sector-specific volatility dynamics. First, it extends the application of CEEMDAN to the aerospace and defense sector, which has been underexplored in volatility studies. Second, it demonstrates the methodological advantages of integrating CEEMDAN with GARCH models, offering a novel approach to analyzing multiscale dynamics in financial time series. The CEEMDAN framework isolates oscillatory components across multiple timescales, enhancing the precision of GARCH models. Results indicate that CEEMDAN offered modest improvements in forecasting accuracy for selected indices and models, without fundamentally altering the best-performing model rankings. This exploratory study contributes to the volatility forecasting literature by demonstrating the contextual applicability of a CEEMDAN-GARCH hybrid framework for complex financial time series. The findings offer preliminary insights into risk management and decision-making for investors, policymakers, and industry participants in a rapidly evolving security landscape.