基于小波-高阶矩模型的投资组合策略——以国际原油市场为例

doi:10.16381/j.cnki.issn1003-207x.2020.10.003

中国管理科学 ›› 2020, Vol. 28 ›› Issue (10): 24-35.doi: 10.16381/j.cnki.issn1003-207x.2020.10.003cstr: 32146.14.j.cnki.issn1003-207x.2020.10.003

基于小波-高阶矩模型的投资组合策略——以国际原油市场为例

朱鹏飞^1,2,3, 唐勇^1,2,3, 钟莉^1,2

1. 福州大学经济与管理学院, 福建福州 350116;
2. 金融数学福建省高校重点实验室(莆田学院), 福建莆田 351100;
3. 福建省金融科技创新重点实验室, 福建福州 350116

收稿日期:2018-08-05 修回日期:2018-11-26 出版日期:2020-10-20 发布日期:2020-11-11
通讯作者: 唐勇(1970-),男(汉族),江苏淮安人,福州大学经济与管理学院,教授,博士生导师,研究方向:金融工程与风险管理,E-mail:tangyong2018@126.com. E-mail:tangyong2018@126.com
基金资助:
国家自然科学基金资助项目（71171056，71573042，71473039）；福建省社科规划重大项目（FJ2017Z006）；福建省自然科学基金资助项目（2017J01518）；金融数学福建省高校重点实验室（莆田学院）开放课题（JR201804）

Portfolio Strategy Based on Wavelet-High Order Moments model-Take the International Crude Oil Markets as An Research Objects

ZHU Peng-fei^1,2,3, TANG Yong^1,2,3, ZHONG Li^1,2

1. School of Economics & Management, Fuzhou University, Fuzhou 350116, China;
2. Key Laboratory of Financial Mathematics(Putian University), Province University, Fujian Putian 351100, China;
3. Fujian Provincial Key Laboratory of finance and technology innovation, Fuzhou 350116, China

Received:2018-08-05 Revised:2018-11-26 Online:2020-10-20 Published:2020-11-11

摘要/Abstract

摘要： 考虑到投资者异质性特征，将极大重叠离散小波变换方法与高阶矩投资组合框架相结合，提出小波-高阶矩投资组合模型，在此基础上提出频域视角下的高频尺度集成方案和时-频域视角下的全尺度集成方案，并遴选出合适的风险偏好特征改进模型，最后进行稳定性检验。基于国际原油市场数据，样本外检验结果表明：相较于对照组，大部分的小波-高阶矩投资组合策略均取得了更优的投资效果，其中集成部分表现最佳，且高频尺度集成方案侧重于提升收益，而全尺度集成方案侧重于降低波动；通过选择合适偏好高阶矩风险的特征，将会明显改善原始小波-高阶矩投资组合策略，且对两个集成方案改良效果最显著；稳健性检验证实了以上结论。

关键词: 小波-高阶矩投资组合策略, 集成方案, 风险特征, 国际原油市场

Abstract: The existing high-order moments portfolio model doesn't take into account the heterogeneity of investors, ignoring the value of multiple time-scales, which is difficult to meet the investors' diversified needs. Therefore, the Maximal Overlap Discrete Wavelet Transform method is combined with the high-order moments portfolio framework to propose the Wavelet-High-order moments portfolio model. By constructing high-order moments portfolios after decomposing the time series, this model can meet the diversified needs of investors in different trading cycles. Then, using the idea of "first decompose and then integrate", a high-frequency scale integration scheme from the frequency domain perspective and a full-scale integration scheme from the time-frequency domain perspective based on the Wavelet-High-order moments portfolio model are proposed. In addition, according to the different risk preference of investors, the appropriate risk features are selected to improve model. Finally, the stability test is carried out. In view of the drastic fluctuation of international crude oil markets, the effect of the portfolio model is tested based on the data of the international crude oil markets. WTI, Brent and Dubai Crude Oil are taken as the research objects. The time spans from October 11th, 2006 to May 16th, 2018, with a total of 2800 trading days after deleting non-common trading days. The data of WTI and Brent are obtained from EIA, and the data of Dubai Crude Oil is obtained from Phoenix Financial Network. The results of the out-of-sample test indicate that, compared with the control groups, most of the Wavelet-High-order moments portfolio strategies have achieved better investment results, with the integration part performing best. And the high-frequency scale integration scheme focuses on improving revenue, while the full-scale integration scheme focuses on reducing fluctuation. By selecting appropriate preference characterristics, the original Wavelet-High-order moments portfolio strategy will be significantly improved, and the improvement effect of the two integration schemes is most significant; The robustness test confirms the above conclusions. This study broadens the research scope of high-order moments portfolio theory, and has theoretical value and practical significance for investors in crude oil market to optimize asset allocation, prevent and resolve market risks.

Key words: Wavelet-High order moments Portfolio strategy, integrated solutions, risk preference features, international crude oil markets

中图分类号:

F830.9

朱鹏飞,唐勇,钟莉. 基于小波-高阶矩模型的投资组合策略——以国际原油市场为例[J]. 中国管理科学, 2020, 28(10): 24-35.

ZHU Peng-fei,TANG Yong,ZHONG Li. Portfolio Strategy Based on Wavelet-High Order Moments model-Take the International Crude Oil Markets as An Research Objects[J]. Chinese Journal of Management Science, 2020, 28(10): 24-35.

参考文献

[1] Ding Zhihua, Liu Zhenhua, Zhang Yuejun, et al. The contagion effect of international crude oil price fluctuations on Chinese stock market investor sentiment[J]. Applied Energy, 2017, 187(1):27-36.
[2] Huang Shupei, An Haizhong, Huang Xuan, et al. Co-movemet of coherence between oil prices and the stock market from the joint time-frequency perspective[J].Applied Energy, 2018, 221(7):122-130.
[3] 潘伟, 王凤侠, 吴婷. 不同突发事件下进口原油采购策略[J].中国管理科学,2016,24(7):27-35.
[4] Jain A, Biswal P C. Dynamic linkages among oil price, gold price, exchange rate, and stock market in India[J].Resources Policy, 2016, 49(9):179-185.
[5] 赵鲁涛, 李婷,张跃军, 等. 基于Copula-VaR的能源投资组合价格风险度量研究[J].系统工程理论与实践,2015, 35(3):771-779.
[6] Markowitz H M. Portfolio selection[J]. Finance, 1952, 7(3):77-91.
[7] 李爱忠, 任若恩, 董纪昌. 基于集成预测的均值-方差-熵的模糊投资组合选择[J].系统工程理论与实践, 2013, 33(5):1116-1125.
[8] Qin Zhongfeng. Mean-variance model for portfolio optimization problem in the simultaneous presence of random and uncertain returns[J]. European Journal of Operational Research, 2015, 245(2):480-488.
[9] 黄金波, 李仲飞, 丁杰. 基于非参数核估计方法的均值-VaR模型[J]. 中国管理科学, 2017, 25(5):1-10.
[10] Zhou Ke, Gao Jiangjun, Li Duan, et al. Dynamic mean-VaR portfolio selection in continuous time[J]. Quantitative Finance, 2017, 17(10):1-13.
[11] Krokhmal P,Palmquist J, Uryasev S. Portfolio optimization with conditional value-at-Risk objective and constraints[J]. Journal of Risk, 2003,(4):11-27.
[12] 张冀, 谢远涛, 杨娟. 风险依赖、一致性风险度量与投资组合-基于Mean-Copula-CVaR的投资组合研究[J]. 金融研究, 2016,(10):159-173.
[13] Liu L, Shi L, Wen Y, et al.Pension fund portfolio based on CVaR-copula[J]. Boletin Tecnico/technical Bulletin, 2017, 55(12):556-563.
[14] Maringer D, Parpas P. Global optimization of higher order moments in portfolio selection[J]. Journal of Global Optimization, 2009, 43(2-3):219-230.
[15] 黄金波, 李仲飞, 丁杰. 基于CVaR的基金业绩测度研究[J].管理评论, 2018, 30(4):20-32.
[16] Lai K K, Yu L, Wang S. Mean-variance-skewness-kurtosis-based portfolio optimization[J]. International Multi-symposiums on Computer & Computational Sciences, 2006, 2(6):292-297.
[17] 蒋翠侠, 许启发, 张世英. 基于多目标优化和效用理论的高阶矩动态组合投资[J]. 统计研究, 2009, 26(10):73-80.
[18] Martellini L, Ziemann V. Improved estimates of higher-order comoments and implications for portfolio selection[J]. Review of Financial Studies, 2010, 23(4):1467-1502.
[19] Nguyen T T. Portfolio selection under higher moments using fuzzy multi-objective linear programming[J]. Journal of Intelligent & Fuzzy Systems, 2016, 30(4):2139-2156.
[20] Chen Wei, Wang Yun, Zhang Jun, et al. Uncertain portfolio selection with high-order moments[J]. Journal of Intelligent & Fuzzy Systems, 2017, 33(3):1-15.
[21] Huang Shupei, An Haizhong, Gao Xiangyun, et al. Time-frequency featured co-movement between the stock and prices of crude oil and gold[J]. Physica A, 2016, 444(15):985-995.
[22] Jammazi R, Reboredo J C. Dependence and risk management in oil and stock markets. A wavelet-copula analysis[J]. Energy, 2016, 107(15):866-888.
[23] Wang Gangjin, Xie Chi, Chen Shou. Multiscale correlation networks analysis of the US stock market:A wavelet analysis[J]. Journal of Economic Interaction & Coordination, 2017, 12(3):1-34.
[24] Jena S K, Tiwari A K, Roubaud D. Comovements of gold futures markets and the spot market:A wavelet analysis[J]. Finance Research Letters, 2018, 24(3):19-24.
[25] 王莹. 全球外汇市场网络结构、货币影响力与货币社区[J].世界经济研究,2018,(2):38-51+134-135.
[26] Zhang X,Lai K K, Wang S. A new approach for crude oil price analysis based on Empirical Mode Decomposition[J]. Energy Economics, 2008, 30(3):905-918.
[27] Li Fangfang,Wang Siya, Wei Jiahua. Long term rolling prediction model for solar radiation combining empirical mode decomposition (EMD) and artificial neural network (ANN) techniques[J]. Journal of Renewable & Sustainable Energy, 2018, 10(1):013704.
[28] Silvo D. The dynamics of return comovement and spillovers between the czech and european stock markets in the period 1997-2010[J]. Finance a úvěr-Czech Journal of Economics and Finance, 2012, 62(4):368-390.
[29] Deora R, Nguyen D K. Time-scale comovement between the Indian and world stock markets[J]. Working Papers, 2013, 29(3):765-776.
[30] Chen M P, Chen W Y, Tseng T C. Co-movements of returns in the health care sectors from the US, UK, and Germany stock markets:Evidence from the continuous wavelet analyses[J]. International Review of Economics & Finance, 2017, 49(3):484-498.
[31] 张世英, 樊智, 郭名媛. 协整理论与波动模型:金融时间序列分析及应用(第3版)[M].北京:清华大学出版社, 2014.
[32] Maharaj E A. Wavelet timescales and conditional relationship between higher-order systematic co-moments and portfolio returns[J]. Quantitative Finance, 2008, 8(2):201-215.
[33] Berger T, Fieberg C. On portfolio optimization:forecasting asset covariances and variances based on multi-scale risk models[J]. The Journal of Risk Finance, 2016, 17(3):295-309.
[34] Li Shiyun. Volatility spillovers in the CSI300 futures and spot markets in China:empirical study based on discrete wavelet transform and VAR-BEKK-bivariate GARCH Model[J]. Procedia Computer Science, 2015, 55(7):380-387.
[35] 熊正德, 文慧, 熊一鹏.我国外汇市场与股票市场间波动溢出效应实证研究-基于小波多分辨的多元BEKK-GARCH(1,1)模型分析[J].中国管理科学,2015,23(4):30-38.
[36] 于孝建, 王秀花, 徐维军. 基于滚动经济回撤约束和下半方差的最优投资组合策略[J]. 系统工程理论与实践,2018,38(3):545-555.
[37] Shiller R J. Irrational exuberance 3rd edition[M]. Princeton:Princeton University Press, 2005.
[38] 叶青, 韩立岩. 基于小波分析研究美国次贷危机在全球股票市场中的传染[J].系统工程,2011,29(5):23-30.
[39] Gunay S. Are the scaling properties of bull and bear markets identical? Evidence from oil and gold markets. Int[J]. Financial Stud, 2014,2(4):315-334.
[40] Berger T, Fieberg C. On portfolio optimization:forecasting asset covariances and variances based on multi-scale risk models[J]. The Journal of Risk Finance, 2016, 17(3):295-309.

基于小波-高阶矩模型的投资组合策略——以国际原油市场为例

Portfolio Strategy Based on Wavelet-High Order Moments model-Take the International Crude Oil Markets as An Research Objects

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 0

[1]	姚银红, 王晓旭, 陈炜, 陈振松. 基于Transformer-LSTM分位数回归的全球股市极端风险溢出研究[J]. 中国管理科学, 2025, 33(8): 1-13.
[2]	吴伟平, 林雨, 金成能, 唐振鹏. 随机市场深度下基于风险管理和交易约束的最优执行问题[J]. 中国管理科学, 2025, 33(8): 14-25.
[3]	田军, 董赞强, 李雅丽. 基于预付账款融资模式的供应链融资策略研究[J]. 中国管理科学, 2025, 33(7): 272-283.
[4]	杨科, 刘鑫, 田凤平. 中国与其他主要新兴市场国家间股市极端风险的跨市场传染[J]. 中国管理科学, 2025, 33(7): 44-53.
[5]	欧阳资生, 周学伟. 中国金融机构系统性风险回测与关联研究[J]. 中国管理科学, 2025, 33(6): 14-26.
[6]	沈根祥, 周泽峰. 拟得分驱动条件异方差自回归极差模型及其实证研究[J]. 中国管理科学, 2025, 33(5): 34-44.
[7]	冯浩原, 吴颉, 于安琪, 郭琨. 杠杆交易会提高股票市场的流动性吗？——基于微观个股层面的实证分析[J]. 中国管理科学, 2025, 33(4): 1-11.
[8]	姚海祥, 刘秋瑜, 杨晓光. 增强还是减弱：新型金融与传统金融行业之间系统性风险溢出[J]. 中国管理科学, 2025, 33(3): 1-12.
[9]	王纲金, 马欣宇, 谢赤. 基于尾部风险溢出网络的全球外汇市场关联性研究[J]. 中国管理科学, 2025, 33(3): 13-23.
[10]	许帅, 邵帅, 何贤杰. 业绩说明会前瞻性信息对分析师盈余预测准确性的影响[J]. 中国管理科学, 2025, 33(3): 34-44.
[11]	孙庆文, 靳玉, 郑立梅, 孙子涵. 企业净资产收益率操纵度量模型与识别方法[J]. 中国管理科学, 2025, 33(2): 1-15.
[12]	胡玉真, 刘建霞. 基于消冗时段网络的集装箱班轮干扰恢复研究[J]. 中国管理科学, 2025, 33(2): 118-130.
[13]	马梦迪, 李烁, 王玉涛. 中国分析师报告有效性研究：特定信息与投资者有限关注[J]. 中国管理科学, 2025, 33(2): 38-49.
[14]	宋诗佳, 田飞, 李汉东. 基于时变极值方法的VaR预测模型以及应用[J]. 中国管理科学, 2025, 33(2): 61-70.
[15]	李星毅, 李仲飞, 李其谦, 刘昱君, 唐文金. 基于机器学习的资产收益率预测研究综述[J]. 中国管理科学, 2025, 33(1): 311-322.