商品期货投资组合与市场收益的尾部相依研究

doi:10.16381/j.cnki.issn1003-207x.2021.1646

Abstract

Abstract:

The diversified allocation of asset portfolios interested many scholars and investors. Additionally， diversifying existing risk factor portfolios is a common strategy for portfolio allocation. However， diversified allocation of factor strategies may not be as advantageous when there is a high correlation between factors. Meanwhile， the Chinese commodity futures market is highly active， and its unique market environment causes different manifestations of risk factors. It is interesting to investigate the correlation between different factors in the Chinese commodity market. It focuses on whether commodity futures factors are applicable in the Chinese futures market and whether the allocation of risk factors can help investors optimize assets.The qpr quantile association model is employed to effectively capture tail dependence. Building upon the work of Li Ruosha et al.， the application of covariates is further expanded upon by integrating the mixed data model MIDAS （Mixed-Data Sampling Model） . Given the relatively short historical data available in China's financial market， the utilization of the MIDAS method becomes crucial. By amalgamating high-frequency daily data， this approach not only enriches the available data information but also enhances the model's fitting capabilities.In this paper， daily high-frequency commodity index and stock index data are used to describe the weekly tail dependence between risk factor returns and market return. Regarding the model： 1） Comparing the basic qpr model， the qpr-MIDAS model can better fit the tail dependence results； 2） the model can realize the diversification return with the allocation of investment portfolios in economic applications. The empirical results show that： 1） there is no obvious spillover relationship between risk factor combinations and market returns except during crises； 2） among all the factors， the momentum factor performs best， with high portfolio return and with low correlation with the market factor； 3） the risk spillover effect negative link with the returns of commodity indexes and stock indexes； 4） Based on our comparative analysis of commodity futures portfolios before and after the change in financial institutions’ access policies， it is determined that the relaxation of policies can help control the intensity of risk spillover. The above research results provide a theoretical reference for Chinese commodity futures investment.The contribution of this paper is that a new tail dependence measurement model qpr-MIDAS is constructed， which can more accurately capture the tail spillover effect under multi-frequency conditions， and explains the advantages of this model from both statistical and economic levels. In addition， based on the perspective of China's commodity futures market， the literature about the law of tail dependence between investment portfolios is extended.This extension serves to offer valuable insights and guidance tailored specifically for Chinese commodity futures investors， thereby enriching their understanding of market dynamics and facilitating more informed investment decisions.

Key words: commodity futures portfolio, tail dependence, MIDAS, quantile association regression model

CLC Number:

F830

Ranran Guo,Wuyi Ye,Xiaoquan Liu,Baiqi Miao. The Tail Dependence Between Commodity Futures Portfolios:Based on qpr-MIDAS Model[J]. Chinese Journal of Management Science, 2024, 32(10): 11-19.

Figures/Tables 8

指数类型	变量	VOL	MOM	FX	LIQ	SK	TS	HP	SPECU	OI
商品指数	$β 0$	0.0711	-0.0282	0.8762^*	-0.0051	0.2426	0.1571	-0.1398	0.1929	0.5733^*
	$β 0$	(0.2113)	(-0.0768)	(3.8811)	(-0.0137)	(0.6915)	(0.5623)	(-0.3749)	(0.602)	(2.3942)
	$β m i d a s$	-5.62^*	3.2196	-4.6552^*	2.7646	3.1705	-9.402^*	-9.2577^*	2.9747^*	0.7137
	$β m i d a s$	(-1.2957)	(0.9811)	(-1.3214)	(0.7102)	(0.8675)	(-4.5554)	(-4.236)	(1.8711)	(0.2909)
	BIC	-5.4538	-5.5203	-5.0843	-5.52	-5.3961	-5.279	-5.4446	-5.3962	-5.2526
股票指数	$β 0$	0.1158	-0.0016	0.8507^*	0.0137	0.1839	0.2677	-0.0549	0.2862	0.5671^*
	$β 0$	(0.3426)	(-0.0043)	(3.616)	(0.0378)	(0.5413)	(0.9265)	(-0.1411)	(0.9424)	(2.3873)
	$β m i d a s$	-2.838	1.5145	-3.2224^*	1.4006	-3.7065^*	-5.4037^*	-5.0227^*	-0.332	-1.13
	$β m i d a s$	(-1.0081)	(0.8021)	(-1.5318)	(0.4306)	(-1.7513)	(-2.4917)	(-2.5664)	(-0.0975)	(-0.5208)
	BIC	-5.4527	-5.5198	-5.0844	-5.5198	-5.397	-5.3024	-5.456	-5.3954	-5.2524

	商品指数为协变量的参数结果
变量	VOL	MOM	FX	LIQ	SK	TS	HP	SPECU	OI
$β 0$	-0.0658	-0.0166	0.8807^*	-0.0487	0.0335	0.3241	0.1521	-0.0107	0.3029
$β 0$	(-0.1616)	(-0.0436)	(3.9700)	(-0.1146)	(0.0829)	(0.9327)	(0.4232)	(-0.0287)	(0.9383)
$β 1$	-0.0701	0.0553	-0.018	-0.0573	0.0502	0.1917	0.0119	-0.0194	0.0089
$β 1$	(-0.5167)	(0.2122)	(-0.1602)	(-0.3251)	(0.2521)	(0.906)	(0.0583)	(-0.1983)	(0.0503)
$β 2$	-0.2893^*	0.0431	-0.1166	0.0509	0.0025	-0.1385^*	-0.0686	-0.3252^*	-0.0025
$β 2$	(-1.8982)	(0.3135)	(-1.0053)	(0.3732)	(0.0216)	(-1.3534)	(-0.4777)	(-3.867)	(-0.0273)
$β 3$	0.0685	0.1035	-0.0333	-0.1416	-0.3130^*	-0.1325	-0.0435	0.0221	0.2330^*
$β 3$	(0.3937)	(0.629)	(-0.3252)	(-0.6517)	(-2.4609)	(-0.7392)	(-0.2338)	(0.2278)	(1.8705)
$β 4$	-0.0047	-0.0009	-0.0636	0.1044	0.0193	-0.0192	0.0479	0.1228^*	0.2158^*
$β 4$	(-0.0241)	(-0.0026)	(-0.6242)	(0.4166)	(0.1504)	(-0.1058)	(0.2691)	(1.3225)	(1.5952)
	BIC值
商品	-5.4362	-5.5054	-5.064	-5.5077	-5.3692	-5.2794	-5.4367	-5.3765	-5.2406
股指	-5.5006	-5.4533	-5.2490	-5.2846	-5.0671	-5.5018	-5.4347	-5.4043	-5.3998

变量	VOL	MOM	FX	LIQ	SK	TS	HP	SPECU	OI
	商品期货指数为协变量的结果
$β 0$	0.8027^*	-0.0042	1.6644^*	0.0358	0.8142^*	1.0929^*	0.7574^*	0.8068^*	0.8176^*
$β 0$	(1.9881)	(-0.0051)	(7.1639)	(0.0588)	(2.0151)	(3.812)	(1.818)	(2.0373)	(2.2544)
$β 2012$	-1.5295^*	-0.0926	-1.9772^*	-0.5435	-1.1294^*	-1.4132^*	-2.2638^*	-1.1283^*	-0.4377
$β 2012$	(-1.5386)	(-0.098)	(-3.0687)	(-0.6782)	(-1.6996)	(-2.0939)	(-2.1447)	(-1.509)	(-0.7795)
$β m i d a s$	0.8959	3.1237	0.1061	4.9533^*	-0.1993	0.9043	2.4582^*	0.9803	-0.1475
$β m i d a s$	(0.7071)	(1.2596)	(0.0897)	(2.1373)	(-0.2258)	(0.3486)	(1.9085)	(0.3168)	(-0.1209)
	股票指数为协变量的结果
$β 0$	0.7479^*	0.1177	1.6666^*	0.3505	0.8096^*	1.0003^*	0.9595^*	0.7589^*	0.7887^*
$β 0$	(2.0213)	(0.1716)	(7.1946)	(0.704)	(1.916)	(3.2672)	(2.6017)	(1.749)	(2.1532)
$β 2012$	-1.4880^*	-0.1602	-1.9824^*	-0.715	-1.1378^*	-1.3461^*	-2.3709^*	-1.1446^*	-0.4379
$β 2012$	(-1.5492)	(-0.1827)	(-3.0727)	(-0.9269)	(-1.7255)	(-2.0271)	(-2.2175)	(-1.5233)	(-0.7853)
$β m i d a s$	-2.3172	1.2391	0.5733	4.0969	1.2819	-2.1520^*	-0.4044	2.3039^*	2.6685
$β m i d a s$	(-0.9766)	(0.551)	(0.3761)	(0.6046)	(0.5473)	(-1.4854)	(-0.3974)	(1.7357)	(0.6742)

References 34

1	梁巨方，韩乾. 商品期货可以提供潜在组合多样化收益吗？［J］. 金融研究， 2017， 446（8）： 129-144.
	Liang J F， Han Q. Can commodity futures provide potential portfolio diversification benefits？［J］. Journal of Financial Research， 2017， 446（8）： 129-144.
2	Asness C S， Moskowitz T J， Pedersen L H. Value and momentum everywhere［J］. The Journal of Finance， 2013， 68（3）： 929-985.
3	Hong Y M， Jun T， Zhou G F. Asymmetries in stock returns： Statistical tests and economic evaluation［J］. The Review of Financial Studies， 2007， 20（5）： 1547-1581.
4	Gao L， Lu L. The volatility behavior and dependence structure of commodity futures and stocks［J］. Journal of Futures Markets， 2014， 34（1）： 93-101.
5	Gao Xin， Nardari F. Do commodities add economic value in Asset Allocation？ New evidence from time-varying moments［J］. Journal of Financial & Quantitative Analysis， 2018， 53（1）：365-393.
6	齐岳，廖科智. 商品金融化背景下商品期货的多样化收益研究［J］. 中国管理科学， 2021， 29（6）： 10-22.
	Qi Y， Liao K Z. Research on the diversification Benefits of commodity futures under the background of commodity financialization［J］. Chinese Journal of Management Science 2021， 29（6）： 10-22.
7	刘映琳，刘永辉，鞠卓. 国际原油价格波动对中国商品期货的影响——基于多重相关性结构断点的分析［J］. 中国管理科学， 2019， 27（2）： 31-40.
	Liu Y L， Liu Y H， Ju Z， The impact of international crude oil price fluctuation on Chinese commodity futures： Based on the correlation structure break point model［J］. Chinese Journal of Management Science， 2019， 27（2）： 31-40.
8	Sadiq M， Lin C Y， Wang K T， et al. Commodity dynamism in the COVID-19 crisis： Re gold， oil， and stock commodity prices， symmetrical？［J］.Resources policy， 2022， 79： 103033.
9	Jiang Y H， Jiang C， Nie H， et al. The time-varying linkages between global oil market and China’s commodity sectors： Evidence from DCC-GJR-GARCH analyses［J］. Energy， 2019， 166： 577-586.
10	Delatte A L， Lopez C. Commodity and equity markets： Some stylized facts from a copula approach［J］. Journal of Banking & Finance， 2013， 37（12）： 5346-5356.
11	杨坤，于文华，魏宇. 基于 R-vine copula 的原油市场极端风险动态测度研究［J］. 中国管理科学， 2017 （8）： 19-29.
	Yang K， Yu W H， Wei Y. Dynamic Measurement of extreme risk among various crude oil markets： Based on R-vine copula［J］. Chinese Journal of Management Science，2017（8）： 19-29.
12	Gaete M， Herrrera R. Diversification benefits of commodities in portfolio allocation： A dynamic factor copula approach［J］. Journal of Commodity Markets， 2023： 100363.
13	Adeleke M A， Awodumi O B. Modelling time and frequency connectedness among energy， agricultural raw materials and food markets［J］. Journal of Applied Economics，2022，25（1）： 644-662.
14	Qiao T S， Han L Y. COVID 19 and tail risk contagion across commodity futures markets［J］. Journal of Futures Markets， 2023， 43（2）： 242-272.
15	Fan J H， Zhang T X. The untold story of commodity futures in China［J］. Journal of Futures Markets， 2020， 40（4）： 671-706.
16	冯玉林，汤珂，康文津. 中国大宗商品期货市场定价机制研究［J］. 金融研究， 2022， 510（12）： 149-167.
	Feng Y L， Tang K， Kang W J. The pricing mechanism in Chinese commodity futures markets［J］. Journal of Financial Research［J］. 2022， 510（12）： 149-167.
17	Campbell J Y， Lettau M， Xu M Y. Have individual stocks become more volatile？ An empirical exploration of idiosyncratic risk［J］. Journal of Finance， 2001， 56（1）：1-43.
18	Ang A， Chen J. Asymmetric correlations of equity portfolios［J］. Journal of Financial Economics， 2002， 63（3）： 443-494.
19	吴吉林，陈刚，黄辰. 中国A， B， H股市间尾部相依性的趋势研究——基于多机制平滑转换混合Copula模型的实证分析［J］.管理科学学报，2015，18（2）：50-65.
	Wu J L， Chen G， Huang C. Long-term dynamic trends in tail dependence of Chinese A， B and H stock markets： Emprical analysis based on multi-regime smoothing transition mixed Copula model［J］. Journal of Management Sciences in China， 2015， 18（2）：50-65.
20	Bergmann D R， Ferreira Savoia J R， De Angelo C F， et al. Portfolio management with tail dependence［J］. Applied Economics， 2018，50（49-51）：5510-5520.
21	Li R S， Cheng Y， Fine J P. Quantile association regression models［J］. Journal of the American Statistical Association， 2014， 109（505）： 230-242.
22	Ye W Y， Luo K B， Liu X Q. Time-varying quantile association regression model with applications to financial contagion and VaR［J］. European Journal of Operational Research， 2017， 256（3）： 1015-1028.
23	Loehlin J C， Nichols R C. Heredity， environment， and personality： A study of 850 sets of twins［M］.Austin， Texas： University of Texas Press， 2012.
24	Ghysels E， Kvedaras V， Zemlys V. Mixed frequency data sampling regression models： The R package midas［J］. Journal of Statistical Software， 2016， 72： 1-35.
25	Ghysels E， Plazzi A， Valkanov R. Why invest in emerging markets？ The role of conditional return asymmetry［J］. The Journal of Finance， 2016， 71（5）：2145-2192.
26	Hu C H， Li Z B， Liu X Y. Liquidity shocks， commodity financialization， and market comovements［J］. Journal of Futures Markets， 2020，40（9）： 1315-1336.
27	李剑，陈烨，李崇光. 金融化与商品价格泡沫［J］. 管理世界， 2018（8）： 84-98.
	Li J， Chen Y， Li C G. Financialization and commodity price bubbles［J］. Journal of Management Word， 2018（8）： 84-98.
28	尹力博，柳依依. 中国商品期货金融化了吗？——来自国际股票市场的证据［J］. 金融研究， 2016， 429（3）： 189-206.
	Yin L B， Liu Y Y. Are Chinese commodity futures markets financialized？ Evidence from international stock markets［J］. Journal of Financial Research， 2016， 429（3）： 189-206.
29	Cheng I H， Xiong W. Financialization of commodity markets［J］. Annual Review of Financial Economics， 2014， 6（1）： 419-441.
30	Kang W J， Tang K， Wang N L. Financialization of commodity markets ten years later［J］. Journal of Commodity Markets， 2023，30： 100313.
31	Pfaff B. Financial risk modelling and portfolio optimization with R［M］.Hoboken， New Jersey： John Wiley & Sons， 2016.
32	Gorton G B， Hayashi F， Rouwenhorst K G. The fundamentals of commodity futures returns［J］. Review of Finance， 2013， 17（1）： 35-105.
33	钟腾，汤珂. 中国商品期货投资属性研究［J］. 金融研究， 2016， 430（4）： 128-143.
	Zhong T， Tang K. Commodity futures as an investment vehicle in China［J］. Journal of Financial Research， 2016， 430（4）： 128-143.
34	Ouyang R， Zhang X. Financialization of agricultural commodities： Evidence from China［J］. Economic Modelling， 2020， 85： 381-389.

描述性统计量	VOL	MOM	FX	LIQ	SK	TS	HP	SPECU	OI	市场收益
均值	0.029	0.276^*	0.100	-0.047	0.128	0.147^*	-0.080	0.081	0.273^*	0.007
方差	2.189	2.831	2.517	2.579	2.141	2.307	2.743	2.604	2.069	1.793
夏普比例	0.013	0.097	0.040	-0.018	0.059	0.063	-0.030	0.031	0.132	0.004

商品	均值	1.1852	1.0355	2.5185	1.0355	1.3314	1.6272	1.1834	1.3314	1.7778
商品	标准差	0.7601	0.3875	0.9931	0.3026	0.3973	3.4385	2.3145	0.6561	0.2371
股指	均值	1.1845	1.0369	2.517	1.0364	1.3301	1.6256	1.1822	1.3312	1.7773
股指	标准差	0.4631	0.2796	1.1673	0.2024	0.7605	1.5653	1.1499	0.0526	0.2644

参数	qpr商品	qpr股指	高斯Copula	反转Gumble	协方差	等权重
	A．样本内
均值	0.2428^*	0.1286^*	0.1003^*	0.1287^*	0.1037^*	0.1009^*
t值	(3.3191)	(5.1089)	(2.7199)	(2.1354)	(2.4469)	(2.1311)
方差	0.7882	1.2319	1.226	1.2181	1.0983	1.5652
夏普比例	0.1154	0.189	0.0968	0.0742	0.0853	0.0758
	B．样本外
均值	0.1298	0.2230^*	0.0548	0.1645	0.1	0.0692
t值	(1.1208)	(1.9443)	(0.5783)	(1.0819)	(0.9028)	(0.8687)
方差	1.258	1.2459	1.0293	1.6517	1.2032	0.8653
夏普比例	0.0952	0.171	0.0435	0.0935	0.0748	0.0684

[1]	Xinyu Wu,Haibin Xie,Chaoqun Ma. Economic Policy Uncertainty and Renminbi Exchange Rate Volatility: Evidence from CARR-MIDAS Model [J]. Chinese Journal of Management Science, 2024, 32(8): 1-14.
[2]	Yangli Guo,Feng MA. Forecasting the Chinese Gold Futures Market Volatility Using Markov-Switching Regime and Mixed Data Sampling Model [J]. Chinese Journal of Management Science, 2024, 32(1): 13-22.
[3]	TAN De-kai, TIAN Li-hui. Is Gold a Safe Haven of the Stock Market?——Based on Dynamic Conditional Correlation Mixed Data Sampling Model [J]. Chinese Journal of Management Science, 2022, 30(10): 14-24.
[4]	WANG Yao-dong, FENG Yan, ZHOU Hua. Does Insurance Play a “Media” Role in the Path of Financial Systemic Risk Contagion?——Empirical Analysis on the Tail Risk Contagion Path of the Chinese Financial Market [J]. Chinese Journal of Management Science, 2021, 29(5): 14-24.
[5]	XU Qi-fa, LIU Shu-ting, JIANG Cui-xia. Portfolio Selection with Conditional Skewness Estimated via MIDAS Quantile Regressions [J]. Chinese Journal of Management Science, 2021, 29(3): 24-36.
[6]	LIU Feng-gen, WU Jun-chuan, YANG Xi-te, OUYANG Zi-sheng. Long-run Dynamic Effect of Macro-economy on Stock Market Volatility Based on Mixed Frequency Data Model [J]. Chinese Journal of Management Science, 2020, 28(10): 65-76.
[7]	ZHU Xiao-qian, LI Jing-yu, LI Jian-ping, CHEN Yi-bin, WEI Lu. An Indicator of Conditional Probability of Crisis for Systemic Risk Measurement [J]. Chinese Journal of Management Science, 2018, 26(6): 1-7.
[8]	XIA Ting, WEN Yue-chun. Does Economic Uncertainty Matter for Stock Market Volatility? An Analysis Based on GARCH-MIDAS [J]. Chinese Journal of Management Science, 2018, 26(12): 1-11.
[9]	JI Qiang, LIU Bing-yue, FAN Ying. Dynamic Dependence Between International Oil, Natural Gas and Exchange Market Based on a New Time-varying Optimal Copula Model [J]. Chinese Journal of Management Science, 2016, 24(10): 1-9.
[10]	YANG Ke, TIAN Feng-ping, LIN Hong. Jump Estimation, Stock Market Volatility Forecasting and Prediction Accuracy Evaluation [J]. Chinese Journal of Management Science, 2013, 21(3): 50-60.
[11]	WU Ji-lin. Nonlinear Relationship of Tail Dependence between Price and Trading Volume in Chinese Stock Markets [J]. Chinese Journal of Management Science, 2012, 20(5): 16-23.

The Tail Dependence Between Commodity Futures Portfolios:Based on qpr-MIDAS Model

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 8

References 34

Related Articles 11

Metrics

Comments

Recommended 0