2008 3 3 :100026788 (2008) 0320025210 CCF ECM 1, 2, 2, 1,2 (11, 100080 ;21, 100080) : (CCF) Hong(2001) ( ECM) Granger. (Dubai) (Tapis) (Minas),,. Hong(2001), 10 %, WTI Brent Granger, WTI., Granger, Hong(2001) ECM. : ; ; ; ; ; ; GARCH : F06411 : A Information spillovers among international crude oil markets - An empirical analysis based on CCF method and ECM LU Feng2bin 1, LI Yi 2, WANG Shuan2hong 2, WANG Shou2yang 1,2 (11Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing 100080, China ; 21School of management of Graduate School of the Chinese Academy of Sciences, Beijing 100080, China) Abstract : Using Hong(2001) s method based on Cross Covariance Function ( CCF) and Error Correction Model ( ECM),we study Granger causality and information spillovers among major global crude oil markets including London, New York, Dubai as well as Tapis and Minas in Southeast Asia. The results show that London and New York futures markets play dominant roles in information spillover. And we find WTI crude oil futures has a slight edge over Brent crude oil futures in information transmission by employing methodology introduced by Hong. In addition, empirical results show Hong s method is more effective than ECM in testing Granger causality and information spillovers. Key words : crude oil futures ; Granger causality ; information spillover ; risk spillover ; cointegration ; ECM; GARCH 1,,,. 2003,2004 8 %, 45147 %,2005 4717 %.,,2005 (WTI) 2004 3618 %, 5617 ;2006 7 14 77103.,.,, GDP,. ( IEA), 10,2006 GDP 018, 018.,,,., :2006209227 : ; :,,, :.
26 2008 3 1.,,,. : Granger ;, CCF Hong (2001), ( ECM), Granger, Hong(2001) ;. 1. 2, Granger,. Brunetti & Gilbert (2000) FIGARCH IPE COMEX, NYMEX IPE, NYMEX. Lin & Tamvakis (2001) GARCH VAR IPE NYMEX,, IPE NYMEX, NYMEX IPE,,NYMEX IPE, IPE,. Hammoudeh et al. (2003) VECM GARCH NYMEX WTI, NYMEX. Hammoudeh & Li (2004) VECM Granger,,.,,. Lin & Tamvakis (2004) ACD IPE NYMEX, NYMEX. Chen et al. (2005),,. Tse & Xiang (2005) NYMEX ( E2Mini), 30 %.,. (2005) Hong(2001) (WTI),,. (2005) VAR Granger, 180 Granger,
3 CCF ECM 27 Granger. 3, (NYMEX) WTI ( IPE) (Brent) (Dubai).,,,2004 27 %. ( ) ( Tapis) ( ICP, - Minas )., (Dubai) (Tapis) (Minas). IPE Brent (NYMEX) WTI Dubai Tapis Minas ; (Brent 2005 4 7 ),WTI WTI Cushing. 2003 1 2 2006 8 25,. r t = 100 (ln ( P t ) - ln ( P t - 1 ) ), P t.,,. 7 947, ( )., WTI WTI,WTICush WTI Cushing,Brent IPE Brent,BrentS Brent,.,.,. Tapis ( ),Minas,Dubai, Brent,NYMEX, 1. IPE Brent WTI, ( 2 3). 1 ( ) Tapis Minas Dubai ( ) 0-1 - 4-8 - 13 2 Brent 1 10 :02-10 :03 18 :02-18 :03 20 :12-20 :13 04 :12-04 :13 2 10 :03-10 :04 18 :03-18 :04 20 :11-20 :12 04 :11-04 :12 3-4 10 :05-10 :06 18 :05-18 :06 20 :09-20 :10 04 :09-04 :10 10 :06-10 :07 18 :06-18 :07 20 :08-20 :09 04 :08-04 :09 2 :00 10 :00 9 :45 17 :45 : Brent 2005 4 7, GMT 1 :00-22 :00. 3 (WTI) 10 :00-14 :30 23 :00-3 :30 15 :15-9 :00 4 :15-22 :00 19 :00-9 :00 8 :00-22 :00 : 2 3 http :ΠΠinfo. china. alibaba. comπnewsπdetailπv4 - d5736631. html. 3,. Brent 2005 4 7 WTI ; IPE,,
28 2008 3 1 :00-22 :00 ( 9 :00-6 :00), 3., Granger, Granger. 311 Hong( 2001) Hong 2001 2 CCF ( ),., 2Granger Q 1 ( M) : Q 1 ( M) = T k 2 ( jπm) ^ 2 ( j) - C 1 T ( k) Π(2 D 1 T ( k) ) 1Π2, j = 1 2Granger Granger, Q 2 ( M) : Q 2 ( M) = T k 2 ( jπm) ^ 2 ( j) - C 2 T ( k) Π(2 D 2 T ( k) ) 1Π2, j = 1- T ^ ( j) j, k ( x) k, M, C 1 T ( k) = C 2 T ( k) = j = 1 (1 - jπt) k 2 ( jπm), D 1 T ( k) = - j ΠT) k j = 1- T(1 2 ( jπm), D 2 T ( k) = j = 1- T j = 1 (1 - jπt) (1 - ( j + 1)ΠT) k 4 ( jπm), (1 - j ΠT) (1 - ( j + 1)ΠT) k 4 ( jπm). Hong(2001),., Granger.,, Daniell, k ( x) Daniell. :, Hong(2001)., GARCH : r t m = c + a t - i r t - i i = 1 + t, t t - 1 N (0, h t ), h t 0 + 2 i t - i + = t t, t t, N (0, h t ) 0 h t. GARCH 4,Ljung2BOX Q. LB (M) 4 GARCH q i = 1 p j = 1 j h t - j, Tapis Minas Dubai Brent BrentS WTI WTICush c 011388 (010584) a 1-010170 (010338) 011498 0 (010590) 010576 1 (010148) 019005 1 (010245) 10 20 111348 [011827] 161954 [015263] 011257 (010686) - 010465 (010341) 011791 (010698) 010563 (010138) 019086 (010223) 414845 [018110 ] 813490 [019730 ] 011211 (010606) - 010685 (010334) 011007 (010621) 010285 (010095) 019436 (010225) 111482 [011759] 141361 [017053] 011790 (010638) - 010654 (010334) 012200 (011167) 010435 (010137) 018993 (010378) 81060 [014276 ] 131597 [017549 ] 011560 (010718) - 010578 (010343) 016197 (012996) 010602 (010211) 018158 (010747) 101149 [012547] 151717 [016123] : (),LB (M) M Ljung2Box, [ ] p. 011390 (010671) - 010493 (010334) 011605 (011065) 010329 (010118) 019284 (010325) 61168 [016284 ] 141162 [017185 ] 011658 (010732) - 011020 (010354) 110576 (016447) 010899 (010377) 017003 (011587) 121150 [011446] 191708 [013495], ( Tapis & Minas) Dubai (Brent) (NYMEX WTI) 4, 4., Q 2 ( M) Granger, Q 1 ( M), 5 6,
3 CCF ECM 29 Q 1 ( M) (Brent WTI )., 8. 5 2Q 2 ( M) 10 % 5 % Tapis ΖMinas Tapis Ζ Dubai Tapis ΖBrent Tapis Ζ WTI Minas Ζ Dubai Dubai ΖBrent Dubai ΖBrentS Dubai Ζ WTI Brent ΖBrentS Brent Ζ WTI BrentS Ζ WTICush 189158 163190 112106 108106 138171 112133 83113 106132 153145 214144 128123 133163 115195 80160 78109 97151 81109 59187 77124 108195 151182 90196 80195 105159 40112 33107 45109 32141 30116 28197 97190 186195 69185 58102 74168 28179 23123 32146 23146 23155 20141 69188 132146 50155 155194 83153 31180 37130 92143 36142 28194 31133 57155 128146 50129 111119 59170 22199 26199 65130 25169 20147 22124 40158 91122 35108 : Ζ ; M 10 20, [ ] p,. 96140 99153 23116 25146 72122 23182 17124 31131 75156 118143 50143 68123 70154 16130 18123 51151 17152 12194 22179 53133 83187 36108 6 2Q 1 ( M) 10 % 5 % Tapis Minas 280173 194199 119184 84129 230161 161141 143196 100160 Tapis Minas - 0128 [0161 ] - 0170 [0176 ] - 0103 [0151 ] 0116 [0144 ] 0136 [0136 ] 0135 [0136 ] - 1113 [0187 ] - 1127 [0190 ] Tapis Dubai 241156 168107 154125 107127 123139 85182 146114 102114 Tapis Dubai 0189 [0119 ] 0154 [0130 ] 2144 [0101 ] 1153 [0106 ] 0138 [0135 ] 1113 [0113 ] 1104 [0115 ] 0151 [0130 ] Tapis Brent 20111 13195 5129 3142 9162 7111 8101 5136 Tapis Brent 150178 104128 56137 38196 38177 26148 27133 18158 Tapis BrentS 47191 33181 5181 5172 18148 13100 11154 9176 Tapis BrentS 76149 52182 24119 16145 17105 11191 9170 7107
30 2008 3 10 % 5 % Tapis WTI 15124 11110 7137 4163 7168 5193 6146 4135 Tapis WTI 149166 103150 43129 29151 49110 33151 32153 22145 Tapis WTICush 15188 11125 2168 1134 [0109 ] 8101 5186 2132 [0101 ] 1136 [0109 ] Tapis WTICush 126105 87115 24186 18107 44185 30190 27139 18131 Dubai Brent 17193 12159 5107 4107 10116 6150 3118 2145 [0101 ] Dubai Brent 153153 106145 44124 30138 45153 31132 33139 23126 Dubai BrentS 45112 32113 3189 5125 28178 19183 15168 11141 Dubai BrentS 80123 55134 41137 28193 14175 10103 10131 7141 Dubai WTI 13147 10155 14138 10122 5144 3146 3128 2172 Dubai WTI 148166 102177 29114 19165 42150 29124 44159 30170 Dubai WTICush 12154 9138 5102 4140 5167 3165 0174 [0123 ] 0167 [0125 ] Dubai WTICush 133153 92150 13164 10106 45104 30180 31189 22120 Brent BrentS 12103 9106 6114 4198 7188 5120 6113 3183 Brent BrentS - 0122 [0159 ] - 0175 [0177 ] - 0186 [0180 ] - 0166 [0174 ] - 0163 [0173 ] - 0158 [0172 ] - 0118 [0157 ] - 0112 [0155 ] Brent WTI - 0105 [0152 ] 0118 [0143 ] - 0153 [0170 ] - 0145 [0167 ] - 0172 [0176 ] - 0149 [0169 ] - 0153 [0170 ] - 0104 [0151 ] Brent WTI - 0172 [0176 ] - 0194 [0183 ] 0130 [0138 ] 0125 [0140 ] 1153 [0106 ] 1131 [0109 ] 1154 [0106 ] 0163 [0126 ] BrentS WTICush 170150 118155 95171 66146 76193 53122 44160 30132 BrentS WTICush 10107 7197 0125 [0140 ] 0180 [0121 ] 6196 5108 4142 3194 WTI WTICush - 0113 [0155 ] 0107 [0147 ] - 0125 [0160 ] - 0135 [0164 ] 1153 [0106 ] 0156 [0169 ] - 0166 [0174 ] - 0172 [0176 ] WTI WTICush 0112 [0145 ] 0102 [0149 ] - 1119 [0188 ] - 1133 [0191 ] - 1105 [0185 ] - 0166 [0175 ] - 0181 [0179 ] - 0139 [0165 ] :. 5,4, p 0,,. 6,, : 1) (Minas & Tapis) Minas Tapis, ;,, Tapis. Tapis Minas 1,
3 CCF ECM 31, Tapis, Tapis,. 2) (Brent &Brent spot), Brent p 0,,,.,. : ; Brent,,Brent. 3) (WTI & WTICush) 5 %, ; 10 %, WTI WTICush 10 % ;,,,,. 4),Tapis.,,,,,.,Tapis Dubai, 1 %., (Dubai),.,,. : 10 %, WTI Brent 10 % 5 %,,, WTI., Tapis Dubai,,. Hong(2001) :,, WTI ;Dubai., ECM, Tapsi Minas,,Dubai Tapis Brent. 312 ECM ECM, ( ),., ECM. ADF, 7 : 1 %,7 Brents,,., EG( Engle2Granger),,,, ECM.,, VECM,, ECM2GARCH : : y t = a + bx t + e t, : y t = c + 0 x t + e t - 1 p + i = 1 i y t - i p + i = 1 i x t - i, u t GRACH( p, q).,, 0 x y, x y y. 7 6 ECM( 8 13),. ( 1 20 Ljing2Box Q ;,, SC ) :
32 2008 3 7 2ADF Tapis Minas Dubai Brent BrentS WTI WTICush - 31603 [01030 ] - 31165-31520 [01038] - 32186-31784 [01018 ] - 33121 : [ ] PP p, :. - 31689 [01024 ] - 32181-31971 [010099] - 32122 8 (Tapis Minas) ECM - 31877 [01013 ] - 31184 M t = 01214 + 01928 T t + e t, gr 2 1 = 01971 T t = - 01113 + 11046M t + e t, gr 2 1 = 01971 M t = 2 10-4 [01557] + 01988 T t - 01038 e t - 1 u t N (0, h t ), gr 2 2 = 01716 h t = 717 10-6 + 01047 u 2 t - 1 + 01898 h t - 1 T t = 417 10-4 + 01725 M t - 01026 e t - 1 [01129 ] u t N (0, h t ), gr 2 2 = 01714 h t = 416 10-6 + 01045 u 2 t - 1 + 01909 h t - 1-31883 [01013] - 34115 : ECM ( [ ] p )., M Minas, Ws Bs ;, e OLS, ; gr 2 1 gr 2 2 R 2.. 9 ECM B t = - 01127 + 11034Bs t + e t, gr 2 1 = 01996 Bs t = 01138 + 01963B t + e t, gr 2 1 = 01996 B t = 412 10-4 + 01777 Bs t - 01057 e t - 1-01014 Bs t - 1 [01043] [01392 ] - 01505 B t - 1 u t N (0, h t ), gr 2 2 = 01629-0116 B t - 2 h t = 511 10-5 + 0124 u 2 t - 1 + 01426 h t - 1 Bs t = - 316 10-4 + 01867 B t - 01084 e t - 1 + 01176 B t - 1 [01130] + 01046 B t - 2 [01015] u t N (0, h t ), gr 2 2 = 01669-01474 Bs t - 1 h t = 6 10-5 + 01242 u 2 t - 1 + 01393 h t - 1-01113 Bs t - 2 [01002 ] 10 ECM W t = - 0102 + 11005 Ws t + e t, gr 2 1 = 01999 Ws t = 01025 + 01993 W t + e t, gr 2 1 = 01999 W t = 115 10-4 + 01856 Ws t - 01484 e t - 1-01002 Ws t - 1 [01608] [01884] - 01107 W t - 3 [01001 ] u t N (0,0100046), gr 2 2 = 01801 Ws t = 515 10-5 + 0194 W t - 0. 551 e t - 1-01015 W t - 1 - [01608 ] [01337 ] 01125 Ws t - 3 u t N (0,0100054), gr 2 2 = 01812 11 (Brent) (WTI) ECM B t = - 0145 + 11103W t + e t, gr 2 1 = 01994 W t = 01429 + 01901B t + e t, gr 2 1 = 01994 B t = 3 10-4 [01078] + 01872 W t - 01043 e t - 1-01018 W t - 1 [01132 ] - W t = 2 10-5 [01915 ] + 01936 B t - 01064 e t - 1-0107 W t - 1 [01033] - 01188 B t - 1-01123 B t - 2 [01003 ] 0112 W t - 2 [01001] u t N (0, h t ), gr 2 2 = 01728 u t N (0, h t ), gr 2 2 = 01731 h t = 218 10-5 + 0141 u 2 t - 1 + 01402 h t - 1 [010001] h t = 214 10-5 + 01591 u 2 t - 1 + 01376 h t - 1
3 CCF ECM 33 12 (Tapis) (WTI) ECM T t = - 0135 + 11093W t + e t, gr 2 1 = 01985 W t = 01374 + 01901 T t + e t, gr 2 1 = 01985 T t = 6 10-4 [01051 ] + 01189 Wt - 01037 e t - 1 + 01524 W t - 1 - W t = 515 10-4 + 01402 T t - 01073 e t - 1-01036 T t - 1 - [01298] [01251] 01278 T t - 1-01165 T t - 2 u t N (0, h t ), gr 2 2 = 01476 + 01089 T t - 3 [01005 ] h t = 714 10-6 + 0106 u 2 t - 1 + 01902 h t - 1 [01017 ] 01218 W t - 1 u t N (0, h t ), gr 2 2 = 01076 h t = 915 10-8 + 01013 u 2 t - 1 + 01986 h t - 1 [01927 ] [01002] 13 (Dubai) (WTI) ECM D t = - 0148 + 11086W t + e t, gr 2 1 = 01972 W t = 01536 + 01895 D t + e t, gr 2 1 = 01972 D t = 417 10-4 + 01173 W t - 0101 e t - 1 + 01558 W t - 1 - [01075] [01027] 01387 D t - 1 u t N (0, h t ), gr 2 2 = 0146 h t = 0100015 + 01237 u2 t - 1-01206 D t - 2-01053 D t - 3 [01067 ] W t = 217 10-4 + 014 D t - 01032 e t - 1 + 01042 D t - 1 - [01560 ] [01001 ] [01231 ] 01257 W t - 1-011 W t - 2 [01012] u t N (0, h t ), gr 2 2 = 01053 h t = 211 10-6 + 010297 u 2 t - 1 + 01965 h t - 1 [01359 ] 5, gr 2 1, 01971.,.,,.,,,.,, ECM. (Tapis Minas) ECM Minas,Tapis 01988,, Tapis Minas Granger, Tapis Minas ; gr 2 2 01716,ECM, Tapis, Tapis Minas Granger. Tapis.,,,, SC,,. ECM,, Brent Granger., Brent, B t - 1 B t - 2, Brent, Bs t - 1, Brent, Granger, Granger.,,. ECM : Granger, Granger ;,Brent WTI Granger, Granger ; Tapis WTI, Tapis WTI, Granger,, Tapis,WTI Tapis, gr 2 2 01476, WTI gr 2 2 ( = 01076), WTI Tapis Tapis WTI,, WTI ;,Dubai WTI Granger, WTI. ECM WTI Granger, Brent WTI Granger. 313
34 2008 3, Hong(2001) ECM, Granger. Hong(2001) Granger, ECM,.,, Granger,Hong(2001) Granger, ECM.,, NYMEX WTI IPE Brent Granger, WTI. Hong (2001), WTI Brent 10 %,,WTI Brent.,.,Lin & Tamvakis(2004) ACD NYMEX,. IPE NYMEX ACD,, ACD, Brent ACD, NYMEX., VECM Granger.,,,.,,.,,,,, Granger,..,,.,,. 4 Hong(2001) ECM Granger, :1) (WTI) (Brent) Granger, ;2) WTI Granger ;3) Brent Granger, Brent Granger, Brent Granger., Hong(2001), WTI WTI Brent,WTI., Granger,Hong(2001) ECM.,, Brent WTI., Brent 2005 4,,.. : [ 1 ] Granger C. Testing for causality : A personal view[j ]. Journal of Economic Dynamics and Control, 1980, 2 : 329-352. [ 2 ] Hong Y. A test for volatility spillover with applications to exchange rates[j ]. Journal of Econometrics, 2001, 103 : 183-224. [ 3 ] Bollerslev T. A generalized autoregressive conditional heteroskedasticity[j ]. Journal of Econometrics, 1986, 31 : 307-27. [ 4 ] Brunetti C, Gilbert C. Bivariate FIGARCH and fractional cointegration[j ]. Journal of Empirical Finance, 2000, 7 : 509-530. [ 5 ] Engle R, Granger C. Co2integration and error correction : Representation, estimation, and testing[j ]. Econometrica, 1987, 55 (2) : 251-276. [ 6 ] Engle R, Russell J. Autoregressive conditional duration : A new model for irregularly spaced transaction data[j ]. Econometrica, 1998, 66 : 1127-1162. [ 7 ] Lin S, Tamvakis M. Spillover effects in energy futures markets[j ]. Energy Economics, 2001, 23 : 43-56. ( 43 )
3 43 : [ 1 ] Axsgter S. Modeling emergency lateral transshipment in inventory systems[j ]. Management Science, 1999, 36 : 1329-1338. [ 2 ] Bowman J P. A multi2echelon inventory stocking algorithm for repairable items with lateral resupply[ Z]. Working Paper, 1986. [ 3 ] Cachon G P. Supply chain coordination with contracts[j ]. Chapter 6 in the Handbooks in Operations Research and Management Science : Supply Chain Management, S. Graves and T. Kok ( Eds. ), North2Holland, 2003. [ 4 ] Dada M. A two2echelon inventory system with priority shipments [ Z]. Working Paper,1985. [ 5 ] Deshpande V, Cohen M A, Donohue K. A threshold inventory rationing policy for service differentiated demand classes [J ]. Management Science,2003,49 : 683-703. [ 6 ] Grahovac J,Chakravarty A. Sharing and lateral transshipment of inventory in a supply chain with expensive low2demand items[j ]. Management Science, 2001, 47 : 579-594. [ 7 ] Lee H L. A multi2echelon inventory model for repairable items with emergency lateral transshipments[j ]. Management Science, 1987,33 : 1302-1316. [ 8 ] McGill J I, Van Ryzin GJ. Revenue management : Research overview and prospects[j ]. Transportation Science, 1999, 33 (2) : 233-256. [ 9 ] Nahmias S, Demmy W S. Operating characteristics of an inventory system with rationing[j ]. Management Science, 1981, 27 : 1236-1245. [10 ] Slay F M. Lateral resupply in a multi2echelon inventory system[ C]ΠΠWorking Note, Logistics Management Instituie, Washington, D. C., 1986. [11 ] Talluri K T, Van Ryzin G J. The Theory and Practice of Revenue Management [M]. Kluwer Academic Publishers, New York, 2004. ( 34 ) [ 8 ] Hammoudeh S, Li H, Jeon B. Causality and volatility spillovers among petroleum prices of WTI, gasoline and heating oil in different locations[j ]. North American Journal of Economics and Finance, 2003, 14 : 89-114. [ 9 ] Lin S, Tamvakis M. Effects of NYMEX trading on IPE brent crude futures markets : A duration analysis[j ]. Energy Policy, 2004, 32 : 77-82. [10 ] Hammoudeh S, Li H. The impact of the Asian crisis on the behavior of US and international petroleum prices [J ]. Energy Economics, 2004, 26 : 135-160. [11 ] Tse Y, Xiang J. Market quality and price discovery : Introduction of the E2mini energy futures[j ]. Global Finance Journal, 2005, 16 : 164-179. [12 ] Bhara R, Hamori S. Causality in variance and the type of traders in crude oil futures[j ]. Energy Economics, 2005, 27 : 527-539. [13 ],,,. [J ]., 2004, 3 (3) : 703-726. [14 ],. [D]. 2005,,2005. [15 ]. [D].,,,2005. [16 ],. [M]. :,2000.