A wavelet based nonlinear ensemble approach for value at risk modeling in metals markets

Subject to shocks worldwide, the metals markets in the era of structural changes and globalization have seen a very competitive and volatile market environment. Proper risk measurement and management in the metals markets are of critical value to investors belonging to different aspects of any economy due to its unique role as important industry inputs to manufacturing process. Although traditional risk management methodologies have worked in the past, the industry is now facing the challenge of rapidly changing market conditions. Markets now demand methodologies that estimate more reliable and accurate VaRs. This paper introduces wavelet analysis to analyze dynamic risk evolution at a finer time scale domain. As wavelet based VaR estimates provide insights into different aspects of the underlying risk evolution, the artificial neural network technique is introduced to find the optimal weight to ensemble these forecasts. Thus this paper proposes wavelet decomposed nonlinear ensemble approach for Value at Risk estimates (WDNEVaR). Empirical studies are conducted using weekly observations in four major metals markets. Performances of the proposed WDNEVaR and the more traditional ARMA-GARCH VaR are evaluated and compared using Kupiec backtesting procedures. Experiment results confirm that VaR estimates produced by WDNEVaR are considerably more reliable and more accurate than traditional methods..