Evaluating the impact of market reforms on Value-at-Risk forecasts of Chinese A and B shares |
| |
| Institution: | 1. 70 Lien Hai RD, Department of Business Management, National Sun Yat-sen University, Kaohsiung 80424, Taiwan;2. 700 University RD Department of Finance, National University of Kaohsiung, Kaohsiung 81148, Taiwan;3. 415 Chien Kung RD, Department of Wealth and Taxation Management, National Kaohsiung University of Applied Sciences, Kaohsiung 80778, Taiwan;1. Department of Money and Banking and Risk and Insurance Research Center, National Chengchi University, 64, Sec. 2, Zhi-Nan Road, Wenshan District, Taipei 11605, Taiwan ROC;2. Department of Money and Banking, National Chengchi University, 64, Sec. 2, Zhi-Nan Road, Wenshan District, Taipei 11605, Taiwan ROC;1. Seoul National University Business School, South Korea;2. Sungkyunkwan University Business School, South Korea;1. Department of Business Management, National United University, Miaoli, Taiwan;2. Department of Banking and Finance, National Chi Nan University, Puli, Taiwan |
| |
| Abstract: | This paper analyses the time-varying conditional correlations between Chinese A and B share returns using the Dynamic Conditional Correlation (DCC) model of Engle Engle, R.F. (2002), “Dynamic Conditional Correlation: A Simple Class of Multivariate Generalized Autoregressive Conditional Heteroskedasticity Models”, Journal of Business and Economic Statistics, 20, 339–350.]. The results show that the conditional correlations increased substantially following the B share market reform, whereby Chinese investors were permitted to purchase B shares. However, this increase in correlations was found to have begun well before the B share market reform. This result has significant implication relating to the structure of the information flow between the markets for the two classes of shares. Value-at-Risk (VaR) threshold forecasts are used to analyse the importance of accommodating dynamic conditional correlations between Chinese A and B shares, and thus reflects the impact of the changes in information flow on the risk evaluation of a diversified portfolio. The competing VaR forecasts are analysed using the Unconditional Coverage, Serial Independence and Conditional Coverage tests of Christoffersen Christoffersen (1998), “Evaluating Interval Forecasts”, International Economic Review, 39, 841–862], and the Time Until First Failure Test of Kupiec Kupiec, P.H., (1995), “Techniques for Verifying the Accuracy of Risk Measurements Models”, Journal of Derivatives, 73–84]. The results offer mild support for the DCC model over its constant conditional correlation counterpart. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
