论文研究-基于时变Copula的金融开放与风险传染.pdf-资料库

weixin_38743506-11789584-4744300845401037866.pdf-第1页.png

第1页 / 共7页

weixin_38743506-11789584-4744300845401037866.pdf-第2页.png

第2页 / 共7页

weixin_38743506-11789584-4744300845401037866.pdf-第3页.png

第3页 / 共7页

weixin_38743506-11789584-4744300845401037866.pdf-第4页.png

第4页 / 共7页

weixin_38743506-11789584-4744300845401037866.pdf-第5页.png

第5页 / 共7页

weixin_38743506-11789584-4744300845401037866.pdf-第6页.png

第6页 / 共7页

weixin_38743506-11789584-4744300845401037866.pdf-第7页.png

第7页 / 共7页

Systems Engineering — Theory & Practice Xó§†¢‚ ¥ã©aÒ: F830.92 [|,4© (úóûŒÆ7Æ,É² 310018) 1 31ò1 4Ï 2011 4 ©?Ò: 1000-6788(2011)04-0778-07 ©zè: A Ä žC Copula7†xD/ Á‡|^žC CopulaïÄ?§e¥Œ †SÌ‡ xD/¯. ^ AR(1)-GJR(1,1)-t.£ãˆÂÇ>S©,±žC SJC Copula£ãÂ ÇÄƒ5,©¥Œ †{ != !F ±9† 2000 1 2010 11ÏxéÄ.¢y²:3?§¥¥Œ †{! =±9F †±‡feƒX,†† eƒ5§ \¥wÍª³;†ˆS ƒ5†±Y². … xD/;7;žC Copula Vol.31, No.4 Apr., 2011 Financial market openness and risk contagion: A time-varying Copula approach WANG Yong-qiao, LIU Shi-wen (College of Finance, Zhejiang Gongshang University, Hangzhou 310018, China) Keywords ﬁnancial contagion; capital market openness; time-varying Copula Abstract The paper proposes a time-varying Copula to study ﬁnancial contagion issues between China mainland and major international stock markets in the opening process of Chinese capital markets. By modeling marginal distributions as AR(1)-GJR(1,1)-t and dependence relations as time-varying SJC Cop- ula, the paper analyzes time-varying co-movements between China mainland and US, UK, Japan, HK stock markets in the interval from Jan 2000 to Nov 2010. The empirical results show that: the lower-tail dependence with US, UK and Japan still stays in a weak level, while the lower tail dependence with HK steadily keeps increasing in the opening process; the upper-tail dependence with all international major stock markets shows a consistent low level. 1ó¥Œ g ±5,‡ƒé4 |Åì=CǑ.7 |Œ© ©.7 |4ž,xŒU3ˆ‡ |D4,ùǑ¥Œ 3æ³7ˆÅ¥ ±3Ïƒ [1].X¥²Åì\.²,¥℄ |ä,k²Œ ®²2‡Õ |.©òïÄ?§¥¥Œ †.Ì‡ ƒxé Ä,dïÄé7 |x››!7i9Ž/57¯‡ †øòäk‡y ¢¿Â.8 ïÄ7 |xD/{Ì‡kƒXê{ [2]!± GARCHǑeõ. [3]!± VAR{ǑÄ:Æ‚=#Ï [1]±94Š [4].ù{Ä,=b ƒ53ž±C, ž©ã,bãSƒ5±C.5û §‚ ÂvFÏ: 2010-05-31 ℄Ï8:<©¬‰ÆïÄÄ7 (10YJC790265);úŽg‰ÆÄ7 (Y7080205);úŽ<©‰:ï ÄÄ/ (7Æ) Šö{0:[|,Æ¬!BÇ,¯7x†7êâ÷ïÄ, E-mail: wangyq@amss.ac.cn;4©,a¬ï Ä,¯7xïÄ.

2.1 SJC Copula [|,:Ä žC Copula7†xD/ 1 4Ï £ã?§¥¥7 |†S7 |xD/†ÄÑïÄ,U‡§¥ ‡5Cz. ©Ì‡^žC Copula{5ïÄ?§e¥ |†SÌ‡ |ƒ xéÄ.Ó§aïÄ¤^{ƒ,d{k‡²w`::^ Copulaê£ã7 | ƒ5,ÓƒXêa{ƒ§U?š‚5ƒ5,Ó GARCHa.† VARa.ƒ§ éÂÇÄ5ÆŠî‚b,U/Ó℄ |š‚5ƒ5†ƒ1Ǒ; žCëêU/£ã?§e¥7 |†.7 |éXÄ5Cz. ©¤^ CopulaêǑé¡ Joe-Clayton (SJC) Copula,§UéB£ãé¡eƒ5† ƒ5.©Xe:1 2©0žC SJC Copulaê9ëê¯,1 3©|^žC SJC Copula£ãy†{ dó²ê!=7ž 100ê!F Nikkei225ê±9 †ðêƒÄƒ,©?1{‡. 2žC Copula Šâ Sklar [5],? Né©Œ±©Ǒ N‡>S©ê†‡ Copulaê.… Xˆ>S©ëY, Copulaê.ù‡ CopulaêÅCƒƒ 5, CopulaêÏdǑ¡Ǒ6ê. Copulaêõé©?CǑ{.X é©,Œk|^Œ{ˆ>S©,2^Œ{ Copulaê. CopulaêŒ±é{/?7 |ƒƒ5,ƒ5£ã‡7 | (),Ä¬å‡ | ( ).X‡7 | CopulaêǑ C(u, v), eƒXê†ƒXê©Ǒ ~„ Copulaêk Copula, t-Copula, Gumbel Copula, Clayton Copula, Frank Copula.ù Copulaê^ £ã7 |xéÄžk²w"€, Copula† t-Copula{Ó7 |šé¡ƒ5, Gumbel Copula{Óeƒ5, Clayton Copula{Óƒ5, Frank Copulaêéƒ5†eƒ5{Ó. Joe[6]Ñ^ Joe-Clayton Copulaê (±e{¡ JC Copula)£ã7 |ƒ5,ê/ ªXe Ó ¡~„ Copulaêƒ, JC CopulaêkwÍ`:,§Œ±Óž£ãƒ5†eƒ 5.éAeƒXê†ƒXêǑ λL = 2−1/γ, λU = 2 − 21/κ. JC CopulaEk‡Ì‡"€,£ãeƒXêƒÓé©ž, JC Copulaêš é¡.ǑŽÑã":, Patton[7]Ñ^é¡ JC Copulaê (±e{¡ SJC Copula)5Ǒx | ƒ5: d SJC CopulaUÓž£ãeƒ5,…U?šé¡eƒ5,Ïd2A^ 7 | 7℄xD/ïÄ. 2.2žC SJC Copula 7 |ƒXŒU²75CzCz,Ïd‡ï žC Copula.£ã 7 |Äš‚5ƒ5 [8].žC Copula.Œa,aëêžC,až C [9−10].©^ëêžCÄ SJC Copula©7 |Äƒ5. tžÅC>Sê©Ǒ f1† f2,éA\È©êǑ F1† F2,©êëê Ǒ θ1† θ2 (XëêÄžC,Œ¤ θ1t† θ2t), SJC CopulažCëêǑ θct = {κt, γt},é CJC (u, v) = 1 −1 −n[1 − (1 − u)κ]−γ + [1 − (1 − v)κ]−γ − 1o−1/γ1/κ CJC (u, v) + CJC (1 − u, 1 − v) + u + v − 1 , λU = lim u→1 , κ ≥ 1, γ > 0 (2) λL = lim u→0 C(u, u) u 1 − 2u + C(u, u) 1 − u CSJC (u, v) = 2 779 (1) (3)

780 ∂u∂v T Xt=1 T Xt=1 T Xt=1 (4) (5) (6) (7) (8) log f2 (x2t; θ2) log f1 (x1t; θ1) , ˆθ2 ≡ arg max θ2 ˆθ1 ≡ arg max θ1 ˆθct ≡ arg max θct f (x1t, x2t; θ1, θ2, θct) = f1(x1t; θ1) · f2(x2t; θ2) · c(F1(x1t; θ1), F2(x2t; θ2); θct) log cF1(x1t; ˆθ1) · F2(x2t; ˆθ2); θct Xó§†¢‚ 1 31ò Ä5ÆŒ±«Ǒ ¥ c(u, v) = ∂C(u,v) Ǒ Copula. žSǑ (x1t, x2t), t = 1, 2, · · · , T .éû Ä5Æëê θ1, θ2† θc?1Œ ž,Œ©?1 [11],1é>S©ëêˆg?1Œ 1é Copulaëê?1Œ 31ž,3 T|Ä: 2T‡ëê,ùŒ1. Patton[7]ïÆ3beƒXê15ÆÄ:?1 CopulaêžCëê, ¥ Λ(·)Ǒ Logisticê Λ(x) = 1/(1 + e−x). \ Λ(·)Ǒy 0 < λL t < 1. ut = F1(x1t; ˆθ1), vt = F2(x2t; ˆθ2).ãz§a ARMA(1, q).,ê q„ 10.d JC Copula‡eƒXê†‡ëêé κt = 1/ log2(2 − λU A,¤±žCëê κt† γtŒ±Ï γt = −1/ log2 λL t )‡í.Ï±ï,ž Cëê κt† γtŒ±Ïéëê ωL, βL, αL, ωU , βU , αUŒ. 3¢y©!±žC SJC Copula.©yê (SSEC)†{ dó²ê (DJIA)!= 7žê (FTSE100)!FF² 225ê (Nikkei225)±9†ðê (HSI)ƒxéÄ X.d kéé,Ïd=±yS/ .{!î³†F.Ì‡² ,©±5.†S‡7¥,†²†Œ²éX;,¤±±ð ê©Œ †† ƒéÄ. 3.1êâ±ˆê±ÂÇŠǑïÄ,ÀžǑ 2000 1 4F 2010 11 1F.d 90 ¥ ¤žÏ, „õ,é §Ǒ,Ä? 4G. ?\#V±5,¥Ï\\ WTO,íÑ QFII† QDII›,í?®ÇU€X„ ℄ |é §Åì\Œ.ÏdÀ 2000 ±êâ5ïÄ?§¥¥ †SÌ‡ ƒxéXä5.ˆÂÇ5Ÿ„ 1.y,S |ÂÇ ÑyÑ²w. Jarque-Bera²ˆêÂÇäk²wš5. 1£ã5 t = Λ ωL + βLλL t = Λ ωU + βU λU |ut−j − vt−j|  |ut−j − vt−j|! q q Xj=1 Xt=1 t < 1, 0 < λU λL λU t−1 + αL · t−1 + αU · 1 q 1 q t Š Jarque-Bera SSEC 0.0013 0.0364 0.0383 4.7614 72 DJIA FTSE100 Nikkei225 0.0000 0.0279 −1.9518 21.8189 8557 −0.0009 0.0267 −0.7468 11.1462 1589 −0.0011 0.0340 −2.2547 25.6194 12324 HSI 0.0008 0.0182 0.0347 7.3024 453

.εt ∼ iid tν µ φ ϕ G A L ν 781 (9) DJIA FTSE100 Nikkei225 HSI r ν σ2 t (ν − 2) t−1 + LI (−εt−1) ε2 t−1 + Aε2 t−1 SSEC 0.0008 0.0480 0.0000 0.9191 0.0492 0.0395 6.1176 0.0018 0.0144 0.0000 0.8990 0.0358 0.0856 13.3770 Rt = µ + φRt−1 + εt σ2 t = ϕ + Gσ2 0.0017 −0.0904 0.0000 0.9120 0.0120 0.1561 8.7773 0.0011 −0.0358 0.0000 0.8165 0.0000 0.2646 9.1396 0.0004 −0.0326 0.0000 0.8828 0.0060 0.0882 6.9261 1 4Ï [|,:Ä žC Copula7†xD/ 3.2>S© Šâ[,æ^ AR(1)-GJR(1,1)-t.ŠˆÂÇ>S©,ä/ªXe I (x)Ǒ«ê, x > 0ž, I (x) = 1, x ≤ 0ž, I (x) = 0.ˆêÂÇ>S©ëê „ 2.éA¤kÂÇ LÑǑ,« |℄öé¡žEk‡A. 2>S©ëê Diehoid [12]‡é[>S©?1‡ :Õ 5 ,Ó© . Ljung- Box ²,3 5%wÍ5Y²eù 5‡žS3gƒ,=ŒǑC†SÕ .|^ Kolmogorov-Smirnov ÄÓ©,= VÇÈ©C†SÄÑ [0,1]!© , 5‡S3 5%wÍ5Y²eÏ .ã² AR(1)-GJR(1,1)-t.Œé£ãˆÂ Ç>S©. 3.3ãy© [13]Ǒ©¥℄ |ª³Œ±ÑÖ℄‡Æ5 . 2001 12¥\\ WTO¿«ì7ÑÖ3ÏSÅ,ù¥℄ |ÑÖ . 2003 71 QFII ,ùX ‚Å℄ö©℄S℄ |, 2006 11S1| QDIIÄ7¤ X¥‚Å℄ö©ª?S℄ |, 2005 7,<¬1®ÇU€¢Sò¥℄ |ÅìíSz.ù‡¯‡é\¯¥℄ |?§äkŒ¿Â,¥©ÅS℄ |5¯‡.Ïd©±ù‡ 5¯‡ŠǑ©.: 2000 ±¥℄ |?§©ǑÊ‡ã,éùÊ‡ãS¥ †.Ì‡ ƒxéX©ƒ± ,±B/£ãxéX3?§¥ÄC z. 3.4 CopulažCëê 3>S©Ä:,|^! 2.2¥1¥{éžC SJC Copula 6‡žCëê?1 Œ.Ǒ©Ä5,é~ê SJC Copula.ëêǑ?1.ëê9.[ ` AIC„ 3. Ä ~ê 3 Copulaëê −2.5295 −2.0490 0.0126 −15.9588 −0.0003 0.0000 −8.6368 −16.0482 −3.8592 1.3835 −2.8254 1.7010 1.1178 −65.6056 −2.3802 4.0845 −8.3223 −11.6201 0.0496 −0.0022 −22.6700 −4.5618 0.0055 0.0055 −6.6865 −0.2058 −0.0011 −6.2898 SSEC-DJIA SSEC-FTSE100 SSEC-Nikkei225 SSEC-HSI ωL βL αL ωU βU αU AIC λU λL AIC 0 0.0426 −9.0154 0.0021 0.0089 −6.2926 0 0.1128 −21.8132 0.0932 0.1313 −48.4748

782 Xó§†¢‚ 1 31ò 3.5eƒ5© y†‡SêžCeƒ5„ã 1.Œ±wÑ¥Œ †{!= ƒeƒ53?§Ê‡ã?3Y²,†{ DJIAeƒXê3 0.05† 2Ä,†= FTSE100eƒXê3 0.01†,=3gˆÅÏǑvkÑy²wÅÄÄÑ. …ã 1Œ²wwÑ:ù‡[eƒ5†²,vkyª³,žC5²w. AICEǑ`²ƒžC SJC Copula,~ SJC CopulaU/[¥Œ †{= ƒƒ5.†=žCeƒXêš~ªC ~ê,Ì‡d eƒXêÄ§¥ ~êXêŒ §Xê.ã©²¥ †{= ƒxéXf,… 3Y²,¥ †{= ÓžÑy4eyVÇé.égˆÅ ƒ 5ïÄkéõ©z,ïÄÑ†©š~C.Ç3† [14]^žC Copula { y¥{ƒgˆÅƒ5k~,Czš~.÷È†W [15]^žCëê t-Copula{ y¥{ 3gˆÅ ƒ5k‡\.Ïd3S℄¥ò℄3 ¥Œ †{= ƒ?1©Ñ℄Uk©Ñ℄x. ¥ †F ƒeƒXê3 0.15†2Ä,Š‡ 0.2,={ ‡Œ.e ƒXêžC5\²w,¿vk²wª³5Cz. AICE²žC SJC Copulaé¥ †F ƒ5k[,¥ †F ƒeƒXêª ž C5.ù²¥ †F ƒxéÄ={ ,¿…3 ÅÄ,¥ † F ÓžÑy4eyŒU5,Œ±ǑX¥Œ℄ |§\,† F xéÄX¿vkwÍ\. ¥Œ †† eƒXê3 ‡ãÑvkyÑª³5,3\ Ñy 6\. 2005 ®ÇU€† 2006 QDII›íÑƒ¥ †† ƒe ƒXêÑyŒÌ.3?§¥¥Œ †† eƒ5yÑŒ žC5,311Êã ÑyŒŠ,žˆ 0.84.¿©`²žC SJC CopulaŒ± /Ó¥Œ †† ƒeƒ5ÄCz,ù‹ AICE.± ²X¥℄ |?§\¯,¥ †† ƒxéÄy5²w,¥ Œ †† ÓžÑy4eyVÇ5Œ. WTO EX WTO EX 0.025 0.02 0.015 0.01 0.005 0 EX EX 0.1 0.08 0.06 0.04 0.02 0 00/01 0.25 0.2 0.15 10/11 10/11 0.1 0.05 0 10/11 00/01 01/12 03/07 00/01 01/12 03/07 1.0 0.8 0.6 0.4 0.2 WTO WTO 01/12 03/07 05/07 06/11 05/07 06/11 05/07 06/11 ¨«žC - -«~ê ã 1y†SÌ‡ |êeƒ5 3.6ƒ5© y†‡SêžCƒ5„ã 2.Œ±wÑ,3¥℄ |?§¥¥ Œ †{!=±9F ƒXêAǑ" (ÑŒ "),²¥ †{= ±9F ÓžŒVÇ4. 2000 ±5,{= ±9F Œvkå¥ Œ,¥ 3 2007 Œ ž{ ÏgˆÅǑ.8 Sé¥ 00/01 01/12 03/07 05/07 06/11 10/11 0

[|,:Ä žC Copula7†xD/ 1 4Ï †SÌ‡ ƒƒ5ïŒÑ^ Copula† t-Copula,Ñéƒ5ïÄ. ƒ{= †F ,¥Œ †† ƒXêƒé‡Œ,Ǒ? Y²,†3 0.1†2Ä.…3¥℄ |§¥ƒé ,¿vkyÑª³.ù ²¥Œ †† ÓžŒVÇǑŒ.†§Fh [16]|^~ê Gumbel CopulaǑx y††ðêƒxéÄ,ÑƒǑ 0.09,†©š~C.d ¥ Œ††;²éX±9õŒúi3/Óž ,Œ²Ñ±9 |ž E ¬é/℄öƒÓÏ,¤±ƒ{= ±9F ,¥Œ †† ƒyÑŒƒ5. 783 ×10 EX 2.5 2 1.5 1 0.5 0 ×10 EX 2.5 2.4 2.3 2.2 2.1 2 10/11 0 0 0.5 0.05 10/11 10/11 00/01 05/07 06/11 01/12 03/07 05/07 06/11 4©^žC SJC Copula.©?§¥¥ †SÌ‡ ÄxéÄ,ž C5Œ±£ãxéXÄ5Cz, SJC CopulaŒ±£ãšé¡eƒ5.¢y ²,¥Œ †{!=±9F ƒÎ±fƒ5†eƒ5, Óž ŒU5Œ,Ä3xD/X;Œ †† éÄ3 \.©ïÄ¤éS7℄kŒëdŠ.d ¥Œ †{!=±9F ƒƒ5,3ù ©Ñ℄Œ±k©Ñ℄x.¥Œ†† ƒ d 3Œƒ5,Aeƒ53¥Œ℄ |§¥kÅ\ª³,3 ¥Œ †† ?1℄ž‡5¿‰4¯‡ ¤E¤℄x,;/ Óž¤E¤ãŒ℄Š. 7xÆ,d ¥Œ †† eƒžC5²w,…yÑÅ\ª³,X^a{öƒ5,é ŒU℄|x. ë©z [1] Hong Y M, Liu Y H, Wang S Y. Granger causality in risk and detection of extreme risk spillover between ﬁnancial markets[J]. Journal of Econometrics, 2009, 150(2): 271–287. [2] King M, Wadhwani S. Transmission of volatility between stock markets[J]. Review of Financial Studies, 1990, 3(1): 5–33. [3] Tse Y K. A test for constant correlations in a multivariate GARCH model[J]. Journal of Econometrics, 2000, 98(1): 107–127. 00/01 01/12 03/07 ×10 2.5 2 1.5 1 EX 05/07 06/11 0.2 0.25 0.15 10/11 00/01 01/12 03/07 05/07 06/11 EX ¨«žC - -«~ê ã 2y†SÌ‡ |êƒ5 00/01 01/12 03/07 0.1

Xó§†¢‚ 1 31ò [4] Longin F, Solnik B. Extreme correlations of international equity markets[J]. Journal of Finance, 2001, 56(2): [5] Sklar A. Fonctions de r´epartition `an dimensions et leurs marges[J]. Publ Inst Statist Univ Paris, 1959(8): 229– 784 649–676. 231. 527–556. 401–423. Gong P, Huang R B. Analysis of the time-varying dependence of foreign exchange assets[J]. Systems Engineering — Theory & Practice, 2008, 28(8): 26–37. [9] Rodriguez J C. Measuring ﬁnancial contagion: A Copula approach[J]. Journal of Empirical Finance, 2007, 14(3): [6] Joe H. Multivariate Models and Dependence Concepts[M]. London: Chapman & Hall, 1997. [7] Patton A J. Modelling asymmetric exchange rate dependence[J]. International Economic Review, 2006, 47(2): [8]÷È,W. ®℄žCƒ5© [J].Xó§†¢‚, 2008, 28(8): 26–37. [13].¥℄ |À [J]..², 2003(3): 67–71. [14]Ç3, .gˆÅ! |x† ƒ5 [J]..², 2010(3): 95–108. [15]÷È,W.gˆÅé¥ Ǒ¢y© ——Ä ¥{ éÄ5© [J]., 2009, 21(2): [16],§Fh.yêðê Copulaƒ5© [J].Xó§, 2006, 24(5): 88–92. 21–32. Gong P, Huang R B. Empirical analysis of sub-prime mortgage crisis’s impacts on Chinese stock market — Based on the interaction between Chinese and American stock markets[J]. Management Review, 2009, 21(2): 21–32. [10] Harvey A. Tracking a changing Copula[J]. Journal of Empirical Finance, 2010, 17(3): 485–500. [11] Patton A J. Estimation of multivariate models for time series of possibly diﬀerent lengths[J]. Journal of Applied Ba S S. Review on the choices of the opening path of Chinese capital markets[J]. World Economics, 2003(3): 67–71. Wu J L, Zhang E H. Subprime mortgage crisis, market risk and dependence between stock markets[J]. World Economics, 2010(3): 95–108. Econometrics, 2006, 21(2): 147–173. International Economic Review, 1998, 39: 863–883. [12] Diebold F X, Gunther T, Tay A S. Evaluating density forecasts with applications to ﬁnancial risk management[J]. Li Y, Cheng X J. Copula dependence analysis between SSEC and HSI[J]. System Engineering, 2006, 24(5): 88–92.

资料库

论文研究-基于时变Copula的金融开放与风险传染.pdf

相关推荐

开发技术

热门标签

最新资料