Forecasting Value-at-Risk using high frequency data: The realized range model Xi-Dong Shao a , Yu-Jun Lian b, ⁎, Lian-Qian Yin a a Jinhe Center for Economic Research, Xi'an Jiaotong University, China b Department of Finance, Lingnan College of Sun Yat-Sen University, Guangzhou 510275, China article info abstract Article history: Received 20 May 2008 Accepted 30 November 2008 Available online 10 June 2009 Current studies on financial market risk measures usually use daily returns based on GARCH type models. This paper models realized range using intraday high frequency data based on CARR framework and apply it to VaR forecasting. Kupiec LR test and dynamic quantile test are used to compare the performance of VaR forecasting of realized range model with another intraday realized volatility model and daily GARCH type models. Empirical results of Chinese Stock Indices show that realized range model performs the same with realized volatility model, which performs much better than daily models. © 2009 Elsevier Inc. All rights reserved. JEL classification: C22 C52 C53 Keywords: VaR Realized range High frequency data 1. Introduction Extreme price movements in the financial markets are rare, but when they occur, they are very mortal. The crisis on Wall Street in 1987 and the crash on Long Term Capital Management have attracted much attention among governments and scholars. Both the risk management of the financial institutions and the risk regulation of the government constitute the core of the current financial market risk. Hence, how to measure the risk value of the financial assets is very important in theory and in practice. Value-at-Risk (VaR) has become a widely used tool in risk measure and management of the financial institutions and the regulations. VaR measures the potential loss of the assets in the advent of a catastrophic event. Current work on VaR focuses on the volatility models based on the GARCH framework, assigning various distributions of the error term of the returns, calculating and predicting VaR. For example, the RiskMetrics model which is widely used by practitioners is just an IGARCH (1, 1) model. Also, the Global Finance Journal 20 (2009) 128–136 ⁎ Corresponding author. Tel.: +86020 84110648. E-mail address: lianyj@mail.sysu.edu.cn (Y.-J. Lian). 1044-0283/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.gfj.2008.11.003 Contents lists available at ScienceDirect Global Finance Journal journal homepage: www.elsevier.com/locate/gfj