Value-at-risk estimation of crude oil price using MCA based transient risk modeling approach Kaijian He a , Kin Keung Lai b, ⁎, Jerome Yen c a Business School, Hunan University of Science and Technology, Xiangtan, Hunan, 411201, China b Department of Management Sciences, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong c Department of Finance and Economics, Tung Wah College, Wylie Road, Kowloon, Hong Kong abstract article info Article history: Received 27 September 2010 Received in revised form 10 January 2011 Accepted 19 January 2011 Available online 27 January 2011 JEL classification: C22 C53 G32 Q47 Keywords: Value at Risk Crude oil Morphological Component Analysis With the increasing level of volatility in the crude oil market, the transient data feature becomes more prevalent in the market and is no longer ignorable during the risk measurement process. Since there are multiple representations for these transient data features using a set of bases available, the sparsity measure based Morphological Component Analysis (MCA) model is proposed in this paper to find the optimal combinations of representations to model these transient data features. Therefore, this paper proposes a MCA based hybrid methodology for analyzing and forecasting the risk evolution in the crude oil market. The underlying transient data components with distinct behaviors are extracted and analyzed using MCA model. The proposed algorithm incorporates these transient data features to adjust for conservative risk estimates from traditional approach based on normal market condition during its risk measurement process. The reliability and stability of Value at Risk (VaR) estimated improve as a result of finer modeling procedure in the multi frequency and time domain while maintaining competent accuracy level, as supported by empirical studies in the representative West Taxes Intermediate (WTI) and Brent crude oil market. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Crude oil serves as an important industrial input and forms the indispensable part of the economy at different levels, which covers a wide range of participants with different interests in the market, such as government, private institutions, etc. This results in numerous factors contributing to the huge level of demand and supply in the crude oil market, such as political and economic events. Thus the price in the crude oil market is influenced by factors with diverse natures and constant shocks, covering each major worldwide events such as the 1973 OPEC oil embargo, Gulf War, 911 and the recent global financial tsunami, etc. The trend is that the market is getting more complex and the volatility level is on the rise. This makes the proper measurement of the risk level essential to the market participants and its smooth operations, which corresponds to the long term interests in research on the accurate measurement of risks from academics, industries and governments (Yang et al., 2002; Plourde and Watkins, 1998). From government perspective, its evaluation and forecasting affect various short-term and long-term decision making process for export policy and national reserves, etc. From academic perspective, it is closely related to some important theoretical issues in the economics and finance fields such as Efficient Market Hypothesis (EMH) and the derivative pricing, etc. From industries perspective, it influences different operational decision making process and the hedging behaviors of individual firms in a wide range of industries such as exporting, manufacturing, transporta- tion, etc. From methodological perspective, in the early development of the market when major shocks are rare and infrequent, academic and industries resort to equilibrium analysis and simple ARMA–GARCH model to measure its risk level, which suffice when the market structure is relatively simpler and could be approximated at the satisfactory level. However, empirical researches fail to reach consensus on their effectiveness and generalizability as they are increasingly challenged with empirical stylized facts such as the fat tail data behavior in recent years and their basic assumptions such as data distributions over fixed time horizon are frequently violated by empirical data. The emerging artificial intelligence approach has produced positive performance due to its data mining effort. However, their performance is largely sensitive to parameters chosen, which lacks rigorous theoretical foundations in the literature and economically viable interpretations (Yu et al., 2008; Zhang et al., 2008). Therefore, alternative risk measurement techniques providing higher reliability and accuracy while being interpretable and tractable are demanded and desired. Semi parametric approaches emerge to fill this gap by analyzing the complicated data features, both its frequency characteristics and time scale characteristics. One approach is the embedding framework to Energy Economics 33 (2011) 903–911 ⁎ Corresponding author. Tel.: + 852 3442 8563; fax: + 852 2788 8560. E-mail addresses: paulhekj@gmail.com (K. He), mskklai@cityu.edu.hk (K.K. Lai), risksolution@gmail.com (J. Yen). 0140-9883/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.eneco.2011.01.007 Contents lists available at ScienceDirect Energy Economics journal homepage: www.elsevier.com/locate/eneco