﻿
 doi:

DOI: 10.3724/SP.J.1249.2018.04432

Journal of Shenzhen University Science and Engineering (深圳大学学报理工版) 2018/35:4 PP.432-440

## Parameter estimation via regime switching model for high frequency data

• LIU Xiangdong 1   JIN Xiaojie 1
• 1.College of Economics, Jinan University, Guangzhou 510632, Guangdong Province, P. R. China

Abstract：
To analyze high-frequency financial data, we use the regime switching method combined with the autoregressive model and volatility replacement model. We perform modelling analysis for high frequency data and use filtering algorithm and Kim smoothing algorithm to perform parameter estimation and prediction. By using the closing prices of each 5 minutes of the 2017-01-03 to 2017-08-02 Shanghai Composite Index, we achieve regime-switching autoregressive model of prediction and speculation, and construct the volatility replacement model. The results show that the Markov high frequency data model is an innovation model and an analysis method with strong theoretical background.

Key words：application of statistical mathematics,high frequency data,regime switching,filtering algorithm,Kim smoothing algorithm,parameter estimation

ReleaseDate：2018-07-26 10:51:04

[1] 朱建平,魏瑾,谢邦昌. 金融高频数据挖掘研究评述与展望[J]. 经济学动态,2011(6):59-62. ZHU Jianping, WEI Jin, XIE Bangchang. Review and prospect of research on financial high frequency data mining[J]. Economic Perspectives, 2011(6):59-62.(in Chinese)

[2] 雷井生,林莎. 基于高频数据的统计套利策略及实证研究[J]. 科研管理,2013,34(6):139-145. LEI Jingsheng, LIN Sha. Statistical arbitrage strategies based on high frequency data[J]. Science Research Management, 2013, 34(6):139-145.(in Chinese)

[3] HANSEN P R, HUANG Z, SHEK H. Realized GARCH:a joint model of returns and realized measures of volatility[J]. Journal of Applied Econometrics, 2012, 27(6):877-906.

[4] 王天一,赵晓军,黄卓. 利用高频数据预测沪深300指数波动率-基于Realized GARCH模型的实证研究[J]. 世界经济文汇,2014(5):18-30. WANG Tianyi, ZHAO Xiaojun, HUANG Zhuo. Forecasting the volatility of Shanghai-Shenzhen 300 index by using high frequency data:an empirical study based on realized GARCH model[J]. World Economic Papers, 2014(5):18-30.(in Chinese)

[5] ZHANG Changyong. Modeling high frequency data using Hawkes processes with power-law kernels[J]. Procedia Computer Science, 2016, 80:762-771.

[6] 李胜歌,张世英. 高频金融数据的两种波动率计算方法比较[J]. 系统管理学报,2007,16(4):426-431. LI Shengge, ZHANG Shiying. Comparative study of two volatility estimation methods of high frequency financial data[J]. Journal of Systems & Management, 2007, 16(4):426-431.(in Chinese)

[7] 唐勇,张世英. 高频金融数据的加权已实现极差波动及其实证分析[J]. 系统工程,2006, 24(8):52-57. TANG Yong, ZHANG Shiying. Weighted realized range-based volatility based on high-frequency data and its empirical analysis[J]. Systems Engineering, 2006, 24(8):52-57.(in Chinese)

[8] 周泽宇. 基于高频数据的跳扩散模型的参数估计[D]. 北京:北方工业大学,2015. ZHOU Zeyu. The jump diffusion models parameter estimation method of high-frequency data[D]. Beijing:North China University of Technology, 2015. (in Chinese)

[9] 徐正国,张世英. 调整"已实现"波动率与GARCH及SV模型对波动的预测能力的比较研究[J]. 系统工程,2004,22(8):60-63. XU Zhengguo, ZHANG Shiying. The comparative research on volatility prediction ability of adjusted realized volatility, GARCH model and SV model[J]. Systems Engineering, 2004, 22(8):60-63.(in Chinese)

[10] 杨科,陈浪南. 基于C_TMPV的中国股市高频波动率的跳跃行为研究[J]. 管理科学,2011,24(2):103-112. YANG Ke, CHEN Langnan. Empirical research on jump behavior of high-frequency volatility based on C_TMPV in Chinese stock markets[J]. Journal of Management Science, 2011, 24(2):103-112.(in Chinese)

[11] 张伟,李平,曾勇. 中国股票市场个股已实现波动率估计[J]. 管理学报,2008, 5(2):269-273. ZHANG Wei, LI Ping, ZENG Yong. Estimating realized volatility of individual stock in Chinese stocks market[J]. Chinese Journal of Management, 2008, 5(2):269-273.(in Chinese)

[12] 沈根祥. 沪深300指数日内跳的Hausman检验[J]. 数理统计与管理,2010,29(4):713-718. SHEN Genxiang. Hausman test for the intraday jumps in Shanghai-Shenzhen 300 index[J]. Journal of Applied Statistics and Management, 2010, 29(4):713-718.(in Chinese)

[13] 唐勇,张伯新. 基于高频数据的中国股市跳跃特征实证分析[J]. 中国管理科学,2013, 21(5):30-39. TANG Yong, ZHANG Boxin. The empirical analysis of jump characteristics of China's stock market based on high frequency data[J]. Chinese Journal of Management,2013, 21(5):30-39.(in Chinese)

[14] BAJGROWICZ P, SCAILLET O, TRECCANI A. Jumps in high-frequency data:spurious detections, dynamics and news[R].[S.l.]:Swiss Finance Institute, 2012.

[15] 汪先珍. 中国股市价格的跳跃行为——基于上证综指高频数据的分析[D]. 厦门:厦门大学,2009. WANG Xianzhen. The jump behavior of China's stock market prices:an analysis with the high frequency data of the SSECI[D]. Xiamen:Xiamen University, 2009.(in Chinese)

[16] HAMILTON J D. A new approach to the economic analysis of nonstationary time series and the business cycle[J]. Econometrica, 1989, 57(2):357-384.

[17] CAI J. A Markov model of switching-regime ARCH[J]. Journal of Business & Economic Statistics, 1994, 12(3):309-316.

[18] HAMILTON J D, SUSMEl R. Autoregressive conditional heteroskedasticity and changes in regime[J]. Journal of Econometrics,1994, 64:307-333.

[19] 佟杰. 基于改进的隐马尔科夫体制转换ARMA-GARCH模型的高频数据预测[D]. 北京:北京理工大学, 2015. TONG Jie. Improved hidden Markov ARMA-GARCH model forecasting in high frequency sample[D]. Beijing:Beijing Institute of Technology, 2015.(in Chinese)

[20] 刘金全, 郑挺国. 利率期限结构的马尔科夫区制转移模型与实证分析[J]. 经济研究, 2006,41(11):82-91. LIU Jinquan, ZHENG Tingguo. Markov regime switching model and empirical analysis of the term structure of interest rates[J]. Economic Research Journal, 2006, 41(11):82-91.(in Chinese)

[21] 杨宝臣, 苏云鹏. SHIBOR市场风险溢价动态特性的机制转换[J]. 系统工程学报, 2012, 27(2):202-207. YANG Baochen, SU Yunpeng. Regime switching in risk premium dynamics of SHIBOR[J]. Journal of Systems Engineering, 2012, 27(2):202-207.(in Chinese)

[22] 王犇. 隐马尔可夫链、马尔可夫状态转换模型及在量化投资中的应用[M]. 北京:清华大学出版社, 2016:110-144. WANG Ben. Hidden Markov chain, Markov switching model, and the applications in quantitative finance[M]. Beijing:Tsinghua University Press, 2016:110-144.(in Chinese)

[23] BOLLERSLEV T. Generalized autoregressive conditional heteroskedasticity[J]. Journal of Econometrics, 1986, 31(3):307-327.

[24] VISSER M P. Garch parameter estimation using high-frequency data[J]. Journal of Financial Econometrics, 2011, 9(1):162-197.

[25] 孙金丽, 张世英. 具有结构转换的GARCH模型及其在中国股市中的应用[J]. 系统工程, 2003, 21(6):87-91. SUN Jinli, ZHANG Shiying. Regime-swithing GARCH in China's stock market[J]. Systems Engineering, 2003, 21(6):87-91.(in Chinese)

[26] GRAY S F. Modeling the conditional distribution of interest rates as regime-switching process[J]. Journal of Financial Economics, 1996, 42(1):27-62.

[27] KLASSEN F. Improving GARCH volatility forecasts with regime-switching GARCH[J]. Empirical Economics, 2002, 27(2):363-394.

[28] 唐勇. 金融资产跳跃检验方法实证比较[J]. 中国管理科学, 2012, 20:290-299. TANG Yong. The empirical comparison among jump tests of financial asset based on high frequency data[J]. Chinese Journal of Management Science, 2012, 20:290-299.(in Chinese)

[29] LIU L, PATTON A J, Sheppard K. Does anything beat 5-minute RV? a comparison of realized measures across multiple asset classes[J]. Journal of Econometrics, 2015, 187(1):293-311.

[30] SILVESTROV D, STENBERG F. A pricing process with stochastic volatility controlled by a semi-Markov process[J]. Communications in Statistics:Theory and Methods, 2004, 33(3):591-608.

[31] ZHU Quanxin. Razumikhin-type theorem for stochastic functional differential equations with Levy noise and Markov switching[J].International Journal of Control, 2017, 90(8):1703-1712.

[32] ZHU Quanxin. Asymptotic stability in the pth moment for stochastic differential equations with Levy noise[J]. Journal of Mathematical Analysis and Applications, 2014, 416(1):126-142.