Forecasting volatility of crude oil markets ☆ Sang Hoon Kang a , Sang-Mok Kang b , Seong-Min Yoon b, ⁎ a Department of Business Administration, Gyeongsang National University, Jinju, 660-701, Republic of Korea b Department of Economics, Pusan National University, Busan, 609-735, Republic of Korea abstract article info Article history: Received 21 January 2008 Received in revised form 3 September 2008 Accepted 10 September 2008 Available online 22 September 2008 JEL classification: Q40 E30 C32 C52 Keywords: Persistence Long memory CGARCH FIGARCH This article investigates the efficacy of a volatility model for three crude oil markets — Brent, Dubai, and West Texas Intermediate (WTI) — with regard to its ability to forecast and identify volatility stylized facts, in particular volatility persistence or long memory. In this context, we assess persistence in the volatility of the three crude oil prices using conditional volatility models. The CGARCH and FIGARCH models are better equipped to capture persistence than are the GARCH and IGARCH models. The CGARCH and FIGARCH models also provide superior performance in out-of-sample volatility forecasts. We conclude that the CGARCH and FIGARCH models are useful for modeling and forecasting persistence in the volatility of crude oil prices. © 2008 Elsevier B.V. All rights reserved. 1. Introduction Modeling and forecasting oil price volatility are important inputs into macroeconometric models, option pricing formulas, and portfolio selection models. For example, current crude oil prices make use of modern financial instruments, including futures contracts, options, and other financial derivatives, and their dynamics are broadly related to economic growth, inflation (Chang and Wong, 2003; Doroodian and Boyd, 2003; Ferderer, 1996; Huang et al., 2005; Huntington, 1998; Lardic and Mignon, 2006; Lee et al., 1995), the price movements of other energy futures contracts (Bhar and Hamori, 2005; Ewing et al., 2006), and other financial assets (Chen and Chen, 2007; Ewing and Thompson, 2007; Pinkdyck, 2001; Sadorsky, 1999, 2000, 2003). In this context, accurate modeling and forecasting of crude oil volatility are of considerable interest to energy researchers and policy makers. Several articles in the energy literature have addressed the modeling and forecasting of crude oil market volatility, using generalized autoregressive conditional heteroskedasticity (GARCH) model and their variants (Adrangi et al., 2001; Cabedo and Moya, 2003; Fong and See, 2002; Giot and Laurent, 2003; Morana, 2001; Narayan and Narayan, 2007; Sadeghi and Shavvalpour, 2006; Sadorsky, 2006). However, there is currently no general consensus on the modeling and forecasting of crude oil volatility, because the standard GARCH models cannot capture persistence in the volatility of crude oil prices. Although the modeling and forecasting of volatility persistence have been widely documented in equity and foreign exchange markets, few studies have thus far analyzed the persistence of long memory in the volatility of crude oil prices (Brunetti and Gilbert, 2000; Tabak and Cajueiro, 2007). Persistence (long memory) means that shocks to conditional variance die at a hyperbolic rate, which is slower than the exponential rate of the decay of shocks in GARCH models (Baillie, 1996). Such a persistent feature is a crucial component of oil risk management, investment portfolios, and the pricing of derivative securities, as its presence is connected closely to the predictability of crude oil volatility (Elder and Serletis, 2008). In response, Baillie et al. (1996) introduced a fractionally integrated GARCH (FIGARCH) model that allows for a factional integrated process in conditional variance, whereas Engle and Lee (1999) developed the component-GARCH (CGARCH) models distinguishing between short- run and long-run persistence of volatility. The principal focus of this paper has been placed on the forecasting of the volatility of three main regional crude oil benchmarks; Brent (North Sea-Europe), Dubai (Persian Gulf), and West Texas Intermedi- ate (WTI) Cushing (US). Recent oil price shocks have been cited as the Energy Economics 31 (2009) 119–125 ☆ This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MOEHRD, Basic Research Promotion Fund) (KRF-2007-521- C00110). ⁎ Corresponding author. Department of Economics, Pusan National University, Jangjeon2-Dong, Geumjeong-Gu, Busan, 609-735, Republic of Korea. Tel.: +82 51 510 2557; fax: +82 51 581 3143. E-mail address: smyoon@pusan.ac.kr (S.-M. Yoon). 0140-9883/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.eneco.2008.09.006 Contents lists available at ScienceDirect Energy Economics journal homepage: www.elsevier.com/locate/eneco