領先指標對台灣景氣趨勢預測能力的探討：非線性因果關係檢定的應用
37
0
0
全文
(2) ¿b −„Vší‘²Vu\£I’×VFÉ·íñ™, 7«ÉË–W\Í% }F)`í −²N™ (¨älN™D°vN™) D −úµ]U Nb, u\\µlW¸þ}×V I’¡5í3b;W Ĥ, % }F)`í®%È−N™, u´?DÄüËã¿«É„Ví −², AÑ%Èç6 \œÉ¸øO×VFÉ·íÿõ …díñíÿu‡ú% }F)`íä lN™5®_,ñCÀ¢‰b, Ç,wúk«É−šíã¿?‰ Bbíl‚à Granger F T|íf$(4Ä‹É[ìj¶ªWõ„«n, êÛ׶MíälN™‰b¸«Éí−š¬ ˙íüæÊ(4íÄ‹É[, Ou¬ íd.·N|−šx.ú˚CÝ(4íÔ4, ¥¶Mí Ä‹É[u(4ìj¶F̶ð|Ví Ñj²,Hí½æ, Bbªø¥‚à Hiemstra and Jones FT|íÝ(4Ä‹É[ìj¶ªW«n, …díõ„!‹6êÛ¶}íälN™‰b° væÊÝ(4íÄ‹É[, ¥_!‹¿Öú„V−‘ªWã¿v, Ý(4WÑFH[ím7u² µÀPF.? eíøá½bÄÖ 7…dõ„!‹FǦ|í’eÝ(4m7, @Œkþ}×V ã¿„V−í‘. ÉœÈ: älN™ −= BDS ì Ý(4Ä‹É[ JEL }éHU: C12, E32. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(3) 1. 1 ‡k ÌuøO¬Ví¾‘ lzlå, Cu¼íI’²µ¸\\µíd•DlW, Ì¿§−‰ í à, ĤÄüËã¿«É„Ví−²£ú−=í~õ −Ø"‚ Yò‚íwì, ˛AÑ%Èç6 \œÉ¸øO×VFÉ·íÿõ 7oÊ 1930 ˝¬, 1ÅÅð%Èû˝ (National Bureau of Economic Research, NBER) ÿ˛% 7øP™ÄD˙å)`®− N™, àJwì1Åí−=~õín‚, %Èùðç66;W NBER íF)`í−N™ ZœÃãí%Èܸl¾j¶, ı?DÄüËwì−í~õ =íVĸšêÞí* ÄÖ 7Ê«ÉË–, ñ‡uJW\Í%È qãº} (˚% }) ì‚Ët0 −²N™(¨ älN™D°vN™) D −úµ]U Nb, TX\\µlW¸þ}×VI’TѲµ¡5¸ ‡ií3bYW, ;W% })`í −²N™ D −úµ]U, ×_,ª}ÑÀ¢4¸õ ”4N™, 1/øÓg,Ù0¸%ÈAÅ0Ñ¡5’e, ÇÕÊwì−²íälN™, ¨Ö7 `¨“hQ¿ÀNb‰0 `¨“©A©~Ì v½É|¨M‰0 ¸ «ÉË–ê ÎÞ û_õ”Þ‰b, £À¢Þ‰b ŒÞX# M1b ‰0 $gNb‰0 ¸ R» ÓgNb‰0 Ĥ, BbªJó]% }Ê)`%ÈN™v, ˛%5¾ƒÀ¢N™Dõ”ÞN™ ª?¥@.°í%Èm7, Ou.ª´wíu, ¤ø)`j¶ºÿˇ%Èܸl¾j¶í! (Š ² ôo, 1993) ku¡ÊÅqÕrÖû˝−=ÔHíd., þt‚à®(4¸Ý(4_«n−= íÉ:4, wì−=í~õ, Cuû˝`§ƒÕʧUí¥@ Ê−:!íõ„û˝jÞ, Aoki (1992) Z‚àÏÏ^£_}&1 n ,úÅÈí−=É:4, 7êÛ3b “Åð −È„.°¥íÛï de Brouwer and Romalis (1996) JŬF)×ß“’e}&1 çß“ −íÉ:˙, 7ç¹ú}5ùß“D1Å−•‘ò~íÉ: Ĥ, Àq ò/† 2” (2002) ‚àÏÏ^£_, }&Åq©Aß|D1 nsÅÈíÅ‚Ì©DÅ‚Ì©Ds‚G|. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(4) 2 cWÑ, 1ô. Pesaran and Shin (1998) T|í§U¥@ƒb (impulse response function), }&.°§UVÄú«Éß|íóÉ˙ Stock and Watson (1989,1991) T|íGÀÄä_ (dynamic factor model), àJ¦³−=2‰bÈíu°š (co-movement),. 1. 7Åq. 튲¸ôo (1993) Z‚àvj¶«n«Éí°v¸älN™ú«É−ã¿í[Û ÊÝ(4_}&íd.2, Hamilton (1989) T|íïªÚ²_ (Markov switching model), ‚à_2íœ0 (transition probability), š−=Ü2F#|íÝú˚4 (asymmetry) ÔH, wwì|í−~õ?D1Å NBER t0í−=n‚Ýó¡ .¤ 5(, rÖç6, à Durland and McCurdy (1994), Ghysels (1994), Lahiri and Wang (1994), Filardo (1994) £ Layton (1998) ‡C(.Ë‚à[ ïªÚ²_2íœ0Ñ ìíc q, UwÓvÈ7‰æ (time-varying), ¤Œkã¿1Åí−~õ Kim and Yoo (1995), Chauvet (1998) £ Kim and Nelson (1998) !¯ Stock and Watson (1989,1991) T|í GÀÄä_¸ Hamilton (1989) íïªÚ²_, ZïªÚ²ÀÄä_ (Markov switching factor model), Î7ªJ°v¦³−=2Ýú˚¸u°šWÑ, yAŠËwìD ã¿1Å−=í~õ 7ÊÅq‚àïªÚ²óÉ_«n«É−=íd.2, ªJ¡ 5Š² ô²Ë 2” (1998), Huang (1999), Chen and Lin (2000a,b) , =} 2” (2001), Chen (2001) ¸ÏDQD2M (2003) ଠíd.·N|−šx.ú˚CÝ(4íÔ4, 7¥¶MíÔ4u(4_F̶ š|Ví, ĤnBVBÖíÝ(4_%â.°íhôiCul¾j¶, J}&,ñv Èåí.ú˚4 7.°k¬ íd.‡ú−=íÉ:4Cu−=í~õwì‡æ í«n, …d‡ú−‘ã¿?‰VTÑ~pí½õ, «nälN™‰bú«É−‘íã¿? ‰ Bbíl‚àf$í(4 Granger Ä‹É[ìj¶, ì®älN™‰bu´uxeäl «É−‘í?‰, Ou−šWÑÖxÝ(4íÔ”, ¥¶}íWÑu(4_F̶…• í, ĤBbQO‚à Hiemstra and Jones (1994) FT|íÝ(4Ä‹É[춪Wõ„} 1 ¥u. Lucas (1977) #|í:−=2í½b,ñ‰bR×v‘íš, D GNP šu°‰íÛï. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(5) 3 & KJüõæÊéOíÝ(4WÑ, †ÈQXM7ïªÚ²_, ƧABc¦_ (threshold autoregressive model), ˲ABc¦_ (smooth transition autoregression model) Ý(4_Êk−šÔ4«n,í£ü4 …díõ„!‹FǦ|í(4CÝ(4Ä‹É [m7, @Œk%Èç6 \œÉ¸øO×Vã¿„V−í‘ …díqñé§à-: ù‚àqíÇ$}&, º¯óÉívÍ’eTX1„, zp«Éä lN™ÄÑ°v¨ÖÀ¢ÞDõ”Þ‰b, 7wÊó°í−ÕG-ª?óæ•‘, 7Öäl‰ bÎ7 v‰b[Û.püÕ, wìí®‰b@vÑß−=‰b (pro-cyclical),. 2. OuÇ$2. [Û|óæí•‘, ºª?¨Awì«É−~õíã¿, ßÞ.Gí à ú Ü…dFS àíõ„_£ìj¶: ¨‚àABc¦ä³_ (vector autoregressive model, VAR), ( 4í Granger Ä‹É[, BDS ì£Ý(4íÄ‹É[ì ûÑõ„!‹, üÑ…d! . 2 «ÉälN™Dd.è 2.1 älN™í)`DŠà −²N™ 2íälN™u% }%âW\Í% } W\Í3lT ‹\¶$lT 2ÛÁW %ûT «É„‰>qF£q\¶Ð óÉ)`œÉTX)7A, ¨ `¨“hQ¿ÀNb ‰0 `¨“©A©~Ì v ½É|¨M‰0 ŒÞX# M1b ‰0 R»Óg Nb‰0 $gNb‰0 ¸ «ÉË–ê ÎÞ þáõ”DÀ¢‰b, w‚šD) `í’eVÄu;W% }ß“−|Œ ÿ“D‹’$l§Ñ ª|¨üq$l~Ñ À¢$l~ Ñ , „‰–”¸Ð ~“N™|ŒD~…, løFÊÕƒí®_áñ}T4í|c, yl_åíhôMD,_~ªœ5‰0, J’e2FhôM×kÉSú˚‰0, J|Û 2 ß−=H[‰b. (àÿ“) Dõ”%ÈAÅ0ÈO£²í?¥?íu°š‘. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(6) 4 ŠMCuɆSÀ‰ 3 ʨälN™D°vN™øÍíóÉ%ÈN™ˇFZAí −²N™ 2, älN™¦ äl%ÈAÅ0‰“, Ĥ%àV¿tcñ%Èí²8$, uã¿„Ví%È•²,|^í $lN™, ?˚ÑlWN™ Ê@à,, øälN™©/,¯ (-±) ú_~eÑ−+` (-±) í™, ¡Víû˝éý, çälN™-± 1∼2 _ì}ª, /šqBýøšíZAáñ-± v, [ý−î Bbø˛9l%¬AÍúb²F’eú`vÈåÇ$ (¡©Ç 1), hô® _‰bvȑǪJêÛ, °v¨ÖÀ¢ÞDõ”Þ‰bíälN™, Êó°í−ÕG-×Öb íN™Ì²,í•‘, 7DÅqÞß„çO£²íÉ[, Ä7‡k2FTƒíß−=Ûï êÞ, OÌ v‰b¸«ÉË–ê ÎÞ ‰0DÅqÞß„ç5È, ÊÇ$,ºÌpéíÉ :4 Q-VÊÇ 2 2, Bbú`|®älN™‰bíAÅ0¸‰0ívÈ‘Ç, ‚àÇ 2 í (a) ÅqÞß„çAÅ0H[õ”%ÈAÅ0íH‰b, Ç 2 2í (b)∼(h) †H[®älN™‰ bí‰0, J®_áñT–4íÇ}&, WàÊ 1973 ¸ 1979 F êísŸù“·œ, .Ø êÛÊÇ 2 í (a) ÅqÞß„çAÅ0D (c) ½É|¨M‰0(e) ©A©~Ì v‰0(f) R»ÓgNb‰0(g) $gNb‰0‰b, ÊøŸù“·œ2, Ì×Ùí«¢, 7×Ö bíälN™6¥ø7−Yòí9õ, 1/Ê 3∼6 _~íäl‚, l(êÞ3§í'òÛï, ó° í, ÊùŸù“·œêÞ5‡, älN™?«¢í²-•‘, 1/oÊ 3∼6 _~‡ÿ˛%¥ø7 −«¢í9õ 7Ê 1987 ∼ 1988 È, À¢4N™[Û|«É−¬ÏíÛï, ¢âk«ÉÎ çÕü|, ‹,ÛWS¦«ÞîM¯MíÕ−\µ, U)×¾Õ’D Ï ÃpÅq, ÅqŒÞX #×ÙAÅ, Ò’Àí U)Åq$ DlËß Òg“Ù, 7éý%È×Û¬37™ Ç“íÕ”, ¥h, õ”Þí‰bÊ¥¨vÈ[ÞM7/šÂò, Ĥ, Ê¥¨‚Èq, À¢Dõ ”‰bºD%ÈAÅ0È, .ø_ím7[ÛWÑ 4 Í7ÿÇ$}&, % })`í®áälN™N˛üõxe¥ø−‘íälŠ?, Ouÿˇ 3 Ì Nb)`j¶~è6¡©â% }%Èû˝}F)í«É−N™~… 4 ÊÀ¢älN™Dõ”älN™m7u´ø_ín,. è6ª¡©ÏDQD2M (2003) íO. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(7) 5 #7‰íçj¶XM®_älN™u´üõälkõ”%ÈAÅ0, Ĥ, …dQOlÿÅqÕ û˝älN™D−=ÈÉ:4íóÉd., lTÀíèDcÜ, 7(yTXœÑÃãí$l j¶íTÑð„. 2.2 óÉd.è A*úÉ˝¬, 1ÅÅð%Èû˝)`®−N™, àJwì1Å−=íVĸšêÞ í*5(, ÊýÉ˝¬, «ÉË–−²N™í)`6â% }ÑÄ@çv«Éí%Èê7 Çá’eíÊÕDcÜ, A¤5(, ZóçÖíç6¤Çáû˝−Ø"Cu«¢íŸÄ −=í~õCuû˝`§ÕV§Uí¥@, 7…dÊ…ü2, ÿ−=2íõ”%ÈAÅD −²N™É:4íû˝‡æ, MøTÅqÕd.nDcÜ ¬ J(4_û˝,ñ%ȉbíd.óçíîó, û˝ä Cu,ñ‰bí²¦yuüIÿ Æ, McGough and Tsolacos (1995) ø,ñ’eT ‘íTÜ(, %âdiÞí’e!Z2) ƒõ”%ÈAÅ0ú “lDlËßxäl^‹ Kunfong and Yong-Yil (1997) }&ÚÅÊ 1970 B 1991 Èí®á,ñN™íÉ:4, 7)ƒª|¨MDŒÞX#ÑälN™, Óg®Ä˘ kr(N™, w2´|¨Mälk%ÈAÅ0ú Hayo (1998) ‚à Granger Ä‹É[ì «nr_ 14 _Åð, ®ÅŒÞX#úß|íüõæÊälÉ[, 7¶MÅð6xeŒÞDß|í¥ —^‹ Martinnez (1999) ‚à3AM}&¶, l¾¦|ú$ Ñ{œx àíú_‰b, ª7Ñ pc¦_2ªW$lR, <ÕíêÛÓ‹¾‘6ÓgNb‰0J£Å¬F)AÅ05(, cñ j„?‰6TòrÖ ÊÝ(4_}&íd.2, Hiemstra and Kramer (1993) ‚àÝ(4 Granger Ä‹É[ «n,$ Ñ{0 (aggregate stock returns) D,ñbÖ (macroeconomic factors), êÛx älN™Š?í,$ Ñ{0D%ÈAÅ0È, Î7(4 Granger Ä‹É[Õ, þ(4_FÌ ¶^|ƒíÝ(4ÔH Jaditz and Jones (1992) ?êÛR»ÓgNb£¾‘6ÓgNbÈ°š xÝ(4Ä‹É[, °vyxeÝ(4í¥—^‹, ¥6éý¤s‰bÊ(4_CrÉhôí|. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(8) 6 /‰búÇø‰bª?xälíÔH, OuʲíÄ‹É[ì, (4_Z̶êr…å‰bÈ í’eÔ4 ÇÕ, Êã¯älN™ (composite leading index, CLI) Dõ”%ÈAÅ0ÈíÉ :4íd., 5 Camacho and Perez-Quiros (2002) ílUà VAR _û˝1Å−2í CLI D GDP 5ÈíÉ:4, 5(ªœ²¾Ë‰c¦_ (vector smooth transition regression, VSTR), !Z²_ (switching regimes model) ¸ Probit _, °š6)ƒ1Åí CLI ü õxäl−•«íN™4<2 BkJälN™û˝Åqû˝−=íd.2, ˜_ (2002) ‡ú$ ŒÞ Ò `‰ Ò − û_À¢ Ò, Uà Granger Ä‹ì §U¥@ƒb >ŒóÉ[b}&j¶, V«n J,û_À¢‰bD%ÈAÅ5ÈíÄ‹É[ älr(É[, ì!‹éýû_À¢‰bú%ÈA Åîxälí Granger Ä‹h1 Š× (2003) ‚àuc¯}& D(4 Granger Ä‹É[ ì‡ú«É$gNb «É£1ÅälN™ D«É£1Å°vN™˛¤È, älCr(8$ªW É:4û˝, )ƒälN™25`¨“hQ¿ÀNb lËß ºC~Þ D`¨“º ©~ T b, % Granger Ä‹ìêÛ„xälÉ[, ˛Ü wälN™5%È<°, @AälN™qñ2 xÎ, Í7, vd.cÿ(4 Granger Ä‹É[¶MT«n, %%ª? I7’eíÝ(4Ô4. 3 õ„j¶ Ê…³, øÝË Ü Granger (1969) (4Ä‹É[ì, Brock,Scheinkman,Dechert and LeBaron (1996) BDS ì, ¸ Baek and Brock (1992) £ Hiemstra and Jones (1994) íÝ( 4Ä‹É[ì 5 ã¯älN™×Ölø®_N™lT™Ä“,. yY®Åv8íÕ”#8.°í½7)„|ã¯älN™, ®Å) „−=N™í Aáñ.`ó°, Wà1ÅíälN™áñÑ Average workweek in manufacturing, Initial claims for unemployment insurance, New orders for consumer goods, Vendor performance, New orders for capital goods, New private housing units authorized by issuance if building permits, Changes in unfilled order for duable goods, Prices of raw materials, Index of stick prices for 500 common stocks, M2 Money stock, Index of consumer expectations 11 _áñ. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(9) 7. 3.1 (4Ä‹É[ì Granger T|Jã¿?‰u´Ó‹íhõVì2‰bÈíÄ‹É[, ì2Ñ: Ly. F(Xt |It−1 ) = F(Xt |(It−1 − Yt−Ly )),. t = 1, 2, . . . ,. (1). Ê (1) 2, {Xt } ¸ {Yt } щ¾(4Óœ¬˙ (bivariate linear stochastic process) FßÞ 50ìb, {It−1 } Ñw t − 1 ‚Èím7Õ¯, ĤBbªe F(Xt |It−1 ) Ñ Xt ‰b#ì {It−1 } Ly. Lx m7Õ¯-í‘Kœ0}º, 7 Yt−Ly ª˚ÑË Ly ˛ÈÅí Yt ‰b5m7Õ¯, D Xt−L u x. ° ¯A {It−1 } m7Õ¯ J (1) .A , Bbªz¬ í Y vÍ’eŒkã¿ÛÊC„Ví X ‰b, ?¹ Y ú X Granger Ä‹É[ ó°í, 5?- : Lx F(Yt |It−1 ) = F(Yt |(It−1 − Xt−L )), x. t = 1, 2, . . . ,. (2). J (2) .A , Bbªz X ú Y Granger Ä‹É[, 6ÿu X vÍ’eŒkã¿ÛÊC„ Ví Y ‰b 7ç (1) ¸ (2) Ì.A , 7²ċÉ[ (two-way causation) íÛï, ?¹ Y ú X Granger Ä‹É[/ X ú Y Granger Ä‹É[, Y ¸ X ívÍ’eŒkã¿Û ÊC„Ví X ¸ Y, Êd.,˚Ñ¥—É[ (feedback) J Granger (1969) ì2ís‰bÈíÄ‹É[, Sim (1980) TX7q¿ìÄ‹É[íj¶, 5?ø_ùjí²¾ABc¦_ (vector autoregression, VAR) : p. Xt =. Yt =. ∑. i=1 r. ∑ k=1. q. αi Xt−i + ∑ β j Yt−j + µt ,. (3). j=1 s. γk Xt−k + ∑ δl Yt−l + νt ,. (4). l=1. (3) ¸ (4) 2, µt νt ÑsóÖ 5ß×á, wœ0}º}Ñ µt ∼ iidN (0, σµ2t ) ¸ νt ∼ iidN (0, σν2t ) , Dkì Y ú X Granger Ä‹É[í™Ìcq¸ú cqª[ýÑ:. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(10) 8 1.. H0A : β j = 0,. j = 1, 2, . . . , q. 2.. H1A : β j 6= 0,. j = 1, 2, . . . , q (β 1 , β 2 , . . . , β j , Býø.ÑÉ). J‡" H0A í™Ìcq, †ª)ƒ Y í¬ vÍm7úÛÊí X éOí(4ã¿?‰, ¹ Y ú X Granger Ä‹É[ °Ü, Ê H0B : γk = 0 í™Ìcq-, J)ƒ‡" H0 í!‹, †ª)ƒ X í¬ vÍm7úÛÊí Y éOí(4ã¿?‰íR, ?¹ X ú Y Granger Ä‹É[; Ê H0A ¸ H0B -, J‡"w™Ìcq, Bbªz²ċÉ[ (bidirectional causality) æÊ. 3.2 BDS ì Ê…³, Bbø 2.1 üíABc¦ä³_ (VAR model) 2íF’eÈí(4É[Î, Í(øF)í{ÏUà BDS ìªWÝ(4É[ì, JìåóÉ4£u´xÝ(4í !Z ¦ã$Í$í}&âlwóÉ&Çá, ÄÑ…xlñq£SšÜXMíÔ4, 7 BDS ì?uJ¤xÍÑ!, qlÑšÓœvÈåF¿RíG 6 Brock, Scheinkman, Dechert and LeBaron (1996) T|í BDS ìu‚àÌ‚b¶ (nonparametric statistical) V}&óÖ 1£Ìø}º (i.i.d) ívÈå’e, ‚à Anderson, Arrrow and Pines (1988) £ Brock (1988) íZ;, y»º Brock, Dechert and Scheinkman (1987) ql|íÌ‚b¶, yªø¥ZG7A BDS ì, JìvÈå’eíÝ(4G_ 5(‚à Gressberger and Procaccia (1983) Fê|íóÉ } (correlaion integral) í– 1, ì2 xt Ñœ0òƒb f FÓœßÞ|ívÈåÑøìb¾, F$A m &ä³ (m-histories) Ñ xtm = (xt , xt+1 , ..., xt+m−1 ), †#ìøÔì m &íóÉ }5là-: Cm.t (ε) = 2. Tm−1. Tm. ∑ ∑. t=1 s=t+1. Iε (xtm , xsm )/(Tm (Tm − 1)). 6 Huang. (5). (1991) ‚àÝ(4GÉ[«nÀ¢ Òíd.2Tƒ: Qïã$Í$ªßÞ|éNëÇßíÓœb WG, Ä7¿R’eöõí…”, ]%%Ê\ÏwÑÓœí8”-, wÝ(4Gí9õ\¿R7.Ag, «wòäí ,ñ‰b (à$ Ñ{0) ª?ßÞéNQ&íã$Í$. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(11) 9 w2,Tm = T − m + 1, Iε (xtm , xsm ) ÑN™ƒb (indicator function) 1 à‹kx m − xsm k ≤ ε, t m m Iε (xt , xs ) = 0 wF8”. kxtm − xsm k =. max. i=0,1,...,m−1. |xt+i − xs+i | < ε. (6). Ĥ, óÉ }u¿¾ÊÔì m &-, ’eÈíÏÊük ε 2F ¯íª0, † T ‚vÈå í BDS $l¾ªì2à-: 1. BDSm,T (ε) = T 2 [Cm,T (ε) − C1,T (ε)]/σm,T (ε). (7). Ê (7) 2,T uš…b, ε uL<²ìíHÜ¡b, σm,T (ε) †Ñú@Ôì m &-íš…™ÄÏ, Ä ¤ BDS ìÑ‚à m ‚Èí’e2, Aú’e5ÏükÔì ε ¸ˇqxˇíÔ4, Jš| ’eF¿R5ÝÓœíG 7. 3.3 Ý(4Ä‹É[ì Brock (1991) wщbÈÝ(4É[æÊv, f$(4_íã¿?‰ø}±Q, J5?øÝ( 4_: Xt = β Yt−L · Xt−M + µt ,. (8). w2, Yt ¸ µt óÖ /wœ0}º}Ñ iidN (0, 1), β Ñw&,¡b, L ¸ M H[ôb , âk Xt ¦²k Yt ¬ íM, Ä7Ê(4í Granger Ä‹É[ì-, øßÞABóÉ[b (autocorrelation) ¸>ŒóÉ[b (cross-correlation) ÌÑÉí#Ï, Ä7Q,7‰bÈíÉ: 4 âkf$í VAR _Ìk(4íã¿?‰, º I7Ý(4í^‹, Ĥ Baek and Brock (1992) ‚àÌ‚b$l¸óÉ }í–1, T|Ý(4 Granger Ä‹É[ì 7 ¬ ʪW. BDS ìv, .â9lsiíY$lܲì¶M¡b, Í7¥ª?²à7._çí¡bM, 7 à7 ì‰ Ä¤ Koˇcenda (2003) T|ZG, fn²ì¡bvsi4˲ì¡bbM, 7uîh˲Ïø–È, é’e…™ ²ì²àœ7í¡bM, øì?‰TB|òíø_bM. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(12) 10 Baek and Brock (1992) ‚à˛ÈóY (spatial dependence) í–1, Ê {X t } ¸ {Yt } 0. ìb˛È2, vÈ t = 1, 2, . . ., ì2 m Ñ X ‰bíäl‚b, 7 Xtm H[â m _äl‚ÈF Ly. Lx Aí²¾, Xt−L Ñ L x _r(‚ÈF Aí²¾, Yt−Ly Ñ Ly _r(‚ÈF Aí²¾, †ªø x. (1) ZŸÑ: Xtm ≡ (Xt , Xt−1 , . . . , Xt+m−1 ), Lx Xt−L ≡ (Xt−L x ,Xt−L x +1 ,...,Xt−1 ) , x. m = 1, 2, . . . ,. t = 1, 2, . . . ,. X L x = 1, 2, . . . ,. Ly. Yt−Ly ≡ (Yt−Ly ,Yt−L x +1 ,...,Yt−1 ) ,. YLy = 1, 2, . . . ,. t = L x+1 , L x+2 , . . . t = Ly+1 , Ly+2 , . . .. (9) (10) (11). J#ì m M, L x ¸ Ly ≥ 1, / e > 0, † Y ¸ X Ý(4 Granger Ä‹É[, /
(13) Ly Ly Lx Lx k < e) − X − Y k < e, kY Pr(kXtm − Xsm k < e
(14) kXt−L s−L t−L s−L y x y x
(15) Lx Lx = Pr(kXtm − Xsm k < e
(16) kXt−L − Xs−L k < e) x x. (12). Lx Ê (12) 2, Pr(·) H[œ0, k · k[ý|׸b (maximun norm), FJ ˝jÑ#ì X t−L x Ly. Ly. Lx ¸ Xs−L Èí|× ×, £#ì Yt−Ly ¸ Ys−L x Èí|× ×Ìük e ím7-, Vã¿ Xtm ¸ x Lx Lx Xsm Èí|× ×ük e ím7-FêÞíœ0, 7 í¬jÑ#ì Xt−L ¸ Xs−L Èí|× x x. ×, Vã¿ Xtm ¸ Xsm Èí|× ×ük e ím7-FêÞíœ0 Ó( Baek and Brock Ñ7jZ (12) íì, ø:¯‘Kœ0í$ ZŸÑÝ:¯‘Kœ0 íªM, w$ Ñ: C1 (m + L x , Ly , e) C3 (m + L x , e) = C2 (L x , Ly , e) C4 (L x , e). (13). w2, C1 (m + L x , Ly , e)/C2 (L x , Ly , e) ¸ C3 (m + L x , e)/C4 (L x , e) }H[ (13) í˝¬j. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(17) 11 ú@í:¯œ05ªM, ©_¶}í:¯œ0ì2à-: Ly. Ly. m+L x m+L x C1 (m + L x , Ly , e) ≡ Pr(kXt−L − Xs−L k < e, kYt−Ly − Ys−Ly k < e ) x x Ly. Ly. Lx Lx C2 (L x , Ly , e) ≡ Pr(kXt−L − Xs−L k < e, kYt−Ly − Ys−Ly k < e) x x m+L x m+L x C3 (m + L x , e) ≡ Pr(kXt−L − Xs−L k < e) x x Lx Lx C4 (L x , e) ≡ Pr(kXt−L − Xs−L k < e) x x. (14) (15) (16) (17). Ê#ì m M, L x ¸ Ly ≥ 1 , / e > 0 -, J (13) A , † Y ¸ X .æÊÝ(4 Granger Ä‹É[ yøóÉ },l¾ (correlation-integral estimators) H (14) ∼ (17) ª): 2 Ly m+Ly m+L x m+L x I(Xt−L , Xs−L , e) · I(Yt−Ly , Ys−Ly , e) ∑ ∑ x x n(n − 1) t<s 2 Ly m+Ly Lx Lx C2 (L x , Ly , e, n) ≡ I(Xt−L , Xs−L , e) · I(Yt−Ly , Ys−Ly , e) ∑ ∑ x x n(n − 1) t<s 2 m+L x m+L x C3 (m + L x , e, n) ≡ I(Xt−L , Xs−L , e) x x n(n − 1) ∑t<s∑ 2 Lx Lx I(Xt−L C4 (L x , e, n) ≡ , Xs−L , e) ∑ ∑ x x n(n − 1) t<s. C1 (m + L x , Ly , e, n) ≡. (18) (19) (20) (21). t, s = max(L x , Ly ) + 1, . . . , T − m + 1, n = T + 1 − m − max(L x , Ly ) Ê (18) ∼ (21) 2, I(X, Y, e) ÑN™ƒb (indicator function), à‹ k X − Y k< e , † I(X, Y, e) = 1, wF†Ñ 0 |(, ‚à (18) ∼ (21) í:¯œ0,lM, BbøªªW (12) í Granger Ä‹É[ì, Ê#ìÔì&˛È m M, L x ¸ Ly ≥ 1, / e > 0 -, / {Xt } ¸ {Yt } sb¯¯ Denken and Keller (1983) cq5ìG, ÿóY4 (weakly dependent), £Rv4 (ergodicity), w}ºÑ Fý: √. C1 (m + L x , Ly , e, n) C3 (m + L x , e, n) a n ∼ N(0, σ2 (m, L x , Ly , e)) − C2 (L x , Ly , e, n) C4 (L x , e, n) . 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics. (22).
(18) 12 w2, σ2 (m, L x , Ly , e) щæb, §ƒ m, L x , Ly , e 5 à. 4 ’ezp£õ„!‹}& 4.1 ’eVÄDzp …dJ% }F)`í −²N™ 2íã¯älN™íþáälN™bÑû˝úï, 8 ¨ `¨“hQ¿ÀNb‰0`¨“©A©~Ì v½É|¨M‰0 ŒÞX# M1b ‰0R»ÓgNb‰0 $gNb‰0 ¸ «ÉË–ê ÎÞ õ”DÀ¢‰ b, ®‰bíŸávÈ‘Ç£AÅ0Ç$cÜkÇ 1 £Ç 2, ’eVÄÑ% }ëjæÜ (http://www.cepd.gov.tw/index.jsp) £©~ì‚|í « É−N™ ‚…, ’eíä0Ñ~’e Ê@à,, øälN™©/,¯ (-±) ú_~eÑ− +` (-±) í™, ¡Víû˝éý, çälN™-± 1∼2 _ì}ª, /šqBýøšíZ Aáñ-±v, [ý−î Bbø˛9l%¬AÍúb²F’eú`vÈåÇ$ (¡©Ç 1) Ѻ¯’eGíõ”ÅqÞß„çAÅ0, 5(z®älN™‰b²A’e (©ú_~ ¦Ìb), TѺ__íš…, š…‚Èâk®‰bí)`–ávÈí.ø_7F.°, <* 1960 ø–, <* 1982 ø–, Oi¢‚ÈîÑ 2003 ù Bbø®älN™‰bí Ì š…‚ÈcÜk[ 1. 4.2 ’eTÜD}& ¬ 'Öíd.îN| (à Nelson and Plosser, 1982), ,ñ‰b£À¢‰b%æÊÝ0ì4í Ô4, JòQªWõ„}&7.5?‰bÔ4v, ª?}ßÞ Granger and Newbold (1974) FT 8 −²N™¨Öã¯älD°vN™,. w2, älN™[W\Í%Èãº}ÑÄ@øÅçv%È꼨5\µ ñ™F)²|5úµ]U, 7®N™5’eVÄâW\Í% } W\Í3lT ‹\¶$lT 2ÛÁW%ûT «É „‰>qF£q\¶Ð óÉ)`œÉTX. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(19) 13 |í ™cc¦ í½æ, ĤʪWÄ‹É[ìí$lR‡, Bblú_‰bªWÀ;ì 9. hô®_‰bvÈåÇ$ (¡©Ç 1 ¸ Ç 2), ªJêÛ®‰bæÊpéívÈ‘, ĤBbJ. ÖvÈ‘í Augmented Dickey-Fuller (ADF) íÀ;ìªW‰b0ì4íì, wì! ‹cÜk[ 2 10 ADF À;ì!‹éý, ®älN™‰bÌ̶‡"xÀ;í™Ìcq, ?¹®å×ÛÝ0 ìÔ4, ĤʪW®älN™åD%ÈAÅ0ÈíÄ‹É[ì5‡, Bb?ú‰bȪWuc ¯ì, ÄÑJ‰bÈxuc¯É[, †®älN™bJJ%¬Ï}TÜ(í’eªWc¦}&, ÖÍfn7™cc¦í½æ, Ou¢ùêÇø_½æ—³5?ƒ‰bȪ?xÅ‚Ì©É[ J ‰bÈæÊuc¯É[v, †Bb‚àÏÏ^£_ªWÄ‹É[ì; J‰bÈ.xuc¯É[, †BbJÏ}í VAR _ªWÄ‹É[ì …d2Bbø‚à Engle and Granger (1987) F T|íù¼¨ì¶ªWuc¯ì, ø¼¨l,luc¯c¦ (co-integrated regression), ù¼¨ìw{Ïáu´Ñ I(1) å, J̶‡" I(1) í™Ìcq, †[ý‰bÈ.xuc¯ É[, ¥5†H[‰bÈxeuc¯É[ BbJ ADF 춪Wù¼¨ì, .J PhilipsOuliaris íìTчiu´æÊ(4uc¯É[íYW, wì!‹cÜk[ 3 â[ 3 2ªø, Î7 `¨“©A©~Ì v Õ, wì®älN™‰b·b²Q§ I(1) í ™Ìcq, ?¹Q§Ìuc¯É[í™Ìcq Bb)ƒ®älN™‰bDõ”ÅqÞß„çÈ.x eÅ‚Ì©É[í!‹1.<Õ, ÄÑ9õ,älN™uŠ?Êkã¿%Ⱥ2ís‚‰“, 7Ê Harvey (1993) ídı6Tƒ, 1Ì#Ëí%ÈÄÖªJXMälN™DÅqÞß„çÈæÊÅ‚ Ì©É[ Ĥ%âuc¯É[ìí!‹, BbÉÊn ©A©~Ì v Dõ”%ÈAÅ 0í(4Ä‹É[ìv, UàÏÏ^£_ªW,l, ‚àÏÏ^£áø‰bÈíÅ‚Ì©É[| c_2, Jfn³Õ½bm7, 7wì®älN™‰bD%ÈAÅ0íÄ‹É[, †‚à‰¾ í²¾ABc¦_ (bivariate vector autoregressive model, BVAR) ªW«n 9 F’e˛l%¬AÍúb² 10 À;ì~¡©. Dickey and Fuller ( 1979, 1981 ). 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(20) 14. 4.3 (4Ä‹É[ì!‹ I Yt Ñ_2ªhôƒí\j„Óœ‰bí‰0, y%â‡øüuc¯É[ì!‹, Bbø õ”%ÈAÅ0 £ ©A©~Ì v, JÏÏ^£_ªW,l, à- Fý: (k) (k) p µ X a a12 θ X t−k + t t = 1 Zt−1 + ∑ 11 (k) (k) νt Yt−k k=1 a21 a22 θ2 Yt wìDõ”%ÈAÅ0ÌÅ‚Ì©É[í®älN™‰0£õ”%ÈAÅ0, †Hp‡øF Üíùjí²¾ABc¦_ (.ÖÏÏ^£á Zt−1 ) 2, wì!‹cÜk[ 4 Ê[ 4 2êÛ, Êõ”älN™‰bjÞ, νÉ|¨M‰0Õ, `¨“hQ¿ÀNb‰ 0 `¨“©A©~Ì v «ÉË–ê ÎÞ H[õ”ÞíälN™‰bÌb²‡"Ì Granger Ä‹É[í™Ìcq, ¥6<âO¥<õ”älN™‰bälkõ”%ÈAÅ0, ¯¯% }²¦í−²N™Œk−AÅíã¿, Ä7ÊÅq%Èêí\µñ™O#7‰í N™4<Ö BbR½É|¨Mú%ÈAÅ.xÄ‹É[íÜâù: ø, ª?uÄÑ«ÉÊc _ÅÒ%È2, ²VrÆOüÇ[í%ÈÔ4, Þúc_ÅÒ\%²í à, |¨Mq§FÅí ˝¬, Ä7Êc_Åqõ”Þíã¿?‰í[Û, ñq§wF%È×ÅF ð, 7̶k}¥øÅq −ší‰“ ù, ½É|¨Mú«É%ÈAÅ0xÝ(4íÄ‹É[, 7¥¶}íÉ[̶ â(4íÄ‹É[ìj¶¿|V ÇÕíu, `¨“hQ¿ÀNb‰0 «ÉË–ê ÎÞ £½É|¨M‰0áñ, ?§õ”%ÈAÅ0í à, 7`¨“hQ¿ÀNb‰0¸«ÉË–ê ÎÞ yDõ”%È AÅ0È, ×Ûó àí¥—É[, ¥ø%ÈíAÅúkõ”Þ?x#Ëí:4 N¬Àí ‹§ŸÜ (accerlation principle) , Ì\ÀPCu¬ÈI’×V, ÊÞú−õßíã‚-Ü, ª?TX7`¨“ØCu¬ÈI’íÓÄ, _U%ÈíAÅõ”ÞAÅí[Û, 7Ê|¨Mí ‰0jÞ, %È−í|êª?U)rÖ ˝Õíß“, ÊÅq−,±í£Þ à-, ª 7ß−=íÛïêÞ, ĤÊÅqI’=1‚Öí×=1-, ¬ ‚í%ÈAÅ, ?ª? Õíß“Ê|¨Mí[Û. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(21) 15 OóúíÊÌ vjÞ, %ÈíAÅUþ}×Víh¨‰Å7, }UAby‹£Uí TCu }Ö*¥? ʬ Éíõ„û˝³1Ìøìíì, 7ÊÜ,íR, wÑAbÊœö¤v, } b²kÖ T (H^‹×k‹ó^‹), 7Êœóv, †b²kÖ*¥ (‹ó^‹×kH^‹), Í7, SvnuË—Dí‹óF), ºuØJ©¾í, uJ, Ê%ÈAÅ0íÌm7-, õÊØ Jú TvÈíÅsTX‰í„p Ĥ, %ÈAÅ0Êú`¨“Ì víã¿, )ƒ7 …d ø_4íÌ(4Ä‹É[í! wŸ, ÊÀ¢ÞälN™íÄ‹É[[Û2, ŒÞX# M1b ‰0¸$gNb‰0úõ”%È AÅ0, ÌæÊ(4í Granger Ä‹É[, 7$ gíšúÅq−šã¿yË p-value Ñ 0.008 íòéOì!‹, ¥@7ÅqÀ¢I’ ÒÊú%ÈAÅ0퉓OBVB½bíä lã¿?‰ 7Ê\ŒÞX# M1b ‰0jÞ, Dõ”%ÈAÅ0?×Q§(4í Granger Ä ‹É[, 6„p7ŒÞb¾uã¿õ”−íGßN™5ø ÊŒÞ\µlW5(, %ÈAÅ0ª? ?ÓŒÞX#¾í 'ò7¹Ö–L, úkþ}×V7k, ªJã‚\ÊlWóÉíŒÞ\µ v, úAÐI’Ü‹d•T|œ‚í‡i, 7ú\7k, Þ@%ÈšíØ"C«¢, ?ªT|£ ü7^í\µ, Ä@ç‡íÅq=1 |(, ÊH[uêg,Ù0íR»ÓgNb‰0, Dõ”%ÈAÅ0?×Û(4íÄ‹É[, éýA…ígšDõ”%ÈíAÅO?¥?퉓É:, 7Êì26 0.001 íòéO ®Ä, Ĥ, ÿÓgšíiõV, R»ÓgNb‰0?uã¿„V%ÈAÅšíälN™‰ b. 4.4 Ý(4Ä‹É[ì!‹ ¬ íd.·N|−šx.ú˚CÝ(4íÔ4, Wà Hamilton (1989), Filardo (1994), Chauvet (1998), Kim and Yoo (1995), Kim and Nelson (1998), Kim and Murray (2002) ,. 11. ¥¶MíÄ‹É[u(4ìj¶F̶ð|Ví Ñj²,Hí½æ, Bbªø¥‚à. 11 É«É−Ô4«níd.,. è6ªJ¡5Š² ô²Ë 2” (1998), ¶È 2” (1999), =}£2. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(22) 16 Baek and Brock (1992) FT|í BDS ì(4_í{Ïáu´‡" Ö /ó°}º(i.i.d.) í™Ìcq, J‡"™Ìcq†[ý‰bȪ?´æÊø<Ý(4íÔ4, 7¥¶MíÔ4uf$( 4Ä‹É[ì¶F̶šƒí QOBb‚à Hiemstra and Jones (1994) FT|íÝ(4 Ä‹É[ìj¶, þtu´?Dð|«ÉälN™bÊã¿−š¬˙2u´æÊÝ(4 Ä‹É[? ®(4ABc¦_í{Ï BDS íì!‹cÜk[ 5, â[2íì!‹ªJhôƒ, ʲ ì.°íp[& (m = 2, 4, . . . , 10) -, ×Öb%â VAR _ Î¥(4j„?‰í®_{Ï M, îb²‡" Ö /ó°}º í™Ìcq, [ý¥<{Ïáíüª?æÊÝ(4ím7 7Ê [ 5 2, Q&íp[Å, wì!‹œb²Q§ Ö /ó°}º í™Ìcq, ¥¸ Ioannis, Ayling and Mahmood (2000) nÛŒÑ{ídı2, Q& (m) íp[ÅœqßÞ{Ïá Ö /ó°}ºí(4’eÔ4í!ó", Ĥ, J Iò&p[˛Èí«n, †ª?}̶ šƒ’eÝ(4íÔH 7ÿ®áñí{ÏVõ, `¨“hQ¿ÀNbʲì&Ñ 4 ‚éOk 1% ]˝®Ä¸ 6 ‚é Ok 5% ]˝®Ä, ‰0½É|¨MÊ 4 B 10 ‚îéOk 1% ]˝®Ä, ŒÞX# M1b Ê 10 ‚éOk 5% ]˝®Ä, 7«ÉË–ê ÎÞ †uÊ 6 ¸ 8 ‚véOk 5% ]˝®Ä, Ĥ¥ û_áñí{Ïáî‡"Ö /}ºó°ícq, 7Bb*¤ûá‰b{ÏáªJhôƒ, 1ÝÊF í&˛ÈÅ2, {Ïáî‡"Ö /}ºó°ícq, OBbwÑÉbø‚m7æÊÝ(4 íÔH, BbZ.?§Îvåª?æÊÝ(4íÔ4, 7Ê‚bœòí{Ïá, ª?óœk‚bQ í{Ïá, Oy#ËíÝ(4m7, Ä7Êœòí&˛È2[Û|wÔ4 ÇÕ, Ê`¨“©A©~Ì v R»ÓgNb‰0¸$gNb‰0í{ÏMî̶‡" Ö /}ºó°í4”, ª?âk’eGíåíš…b.—, Ä7̶k}éýwÝ(4íÔ 4, OBbYͪW Baek and Brock (1992) íÝ(4Ä‹É[ì, V«˝®älN™íÝ(4 Ä‹É[ ” (2001), ÏDQ£2M (2003), Huang (1999), Chen and Lin (2000a, b) . 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(23) 17 ʪWÝ(4Ä‹É[ì‡, BbÛ9l²ìäl‚b (m), ʇø(4íÄ‹É[ì2 í VAR _, BbªJ®8ÑĆTѲ¦äl‚bíYW, OÊ‚àÌ‚bj¶ê|íÝ( 4_ ³, Ê]ô‚b L x Ly £ ¡b e ²¦,, ºÌd.TX„ í²¦j , ĤBbY f$Ý(4 Granger Ä‹É[ìíd., ¡5 Hiemstra and Jones (1994) íqìj , 12 1 ;W’eÔ4Dìíêc4, …døäl‚b (m) qìÑ 1, ]ô‚b L x = Ly qìÑ 1,...,7, 7 ¡b e †qìÑ 0.5σ, w2 σ=1, 7Ý(4Ä‹É[ì!‹cÜk[ 5 [ 6 .°k5‡í)§j , øÝ(4Ä‹É[ìí‰bßå½h§, zälN™–}Ñõ ”Þ‰bDÀ¢Þ‰b, Êõ”Þ‰bíì!‹jÞ, hQ¿ÀNb‰0Êr(ø‚vúkõ” GDP AÅ0æÊÝ(4Ä‹É[, 7ÊwFr(‚†.éO, BbªJR, välN™Öú GDP AÅ0Ý(4íj„?‰, ĤæÊÝ(4íÄ‹É[, 9^‹N˛1.#Ë ÊÌ váír (þ‚2, Îý‚.éOÕ, w쮂îD GDP AÅ0æÊÝ(4íÄ‹É[ ÇÕʽÉ|¨ M‰0jÞ, w 3 B 6 ‚?D GDP AÅ0æÊÝ(4íÄ‹É[, Ÿl”O½É|¨Mu´_¯ TÑã¿−²íN™, ºÊÝ(4íÄ‹É[ì2, AŠíšƒwÝ(4Ä‹É[Ô4 ¥ h«ÉË–ê ÎÞ ‰0í®‚î.éO, 7ú GDP AÅ0.æÊÝ(4íj„?‰, Ĥ Î7ê ÎÞ ‰0áñÕ, ×Öbíõ”4íN™‰búkõ”%ÈAÅ0æÊÝ(4ím 7 ÇÕ, õ” GDP AÅ0úÌ v ½É|¨MD«ÉË–ê ÎÞ ‰0?Ý(4í äl^‹, 7DÌ v¸½É|¨M‰0yæÊÝ(4í¥—^‹, ¥_!‹D(4Ä‹É[1 Ìø_íÉ:, 7/ÀõhQ¿ÀNb‰0Ê(4_2Dõ”%ÈAÅ0O(4í¥—^‹ Võ, välN™‰b‚à(4_¹ªJøw’em7¾¦êH, ĤÝ(4_1Ì”ì5Ý( 4m7ªX¾¦ QOBbhôÀ¢ÞíN™‰bí[Û, R»ÓgNb‰0ír( 6 ¸ 7 ‚úõ” GDP A Å0Ê 10% ]˝®Ä-éOæÊÝ(4Ä‹É[ $gNb‰0ír( 3 ¸ 7 ‚DŒÞX#‰ 12 Hiemstra. and Jones (1994) Ên$ g¾íÝ(4Ä‹É[uøäl‚b (m) qìÑ 1, ]ô‚b L x = Ly qìÑ 1,...,8, 7 ¡b e †qìÑ 1.5σ. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(24) 18 0í 135 ‚?ó°í!‹, 7õ” GDP AÅ0D$gNb‰0È, †×ÛÝ(4í¥— Ûï ÿÀ¢ŒÞÞVõ, älN™úkõ”%ÈAÅ0?xÝ(4í^‹, OÌÿr(‚bC uéO4, Dõ”Þí®‰bóª, wÝ(4^‹î.àõ”Þ‰bí[Û Ä¤ÿÝ(4Ä‹É[ íiVõ, wÝ(4Ô”pé.àõ”Þ‰bí[ÛV)#Ë, Crúk−Ý(4¶}šíã ¿, õ”Þ‰bª?}uy7íälN™‰b ;W,Hì!‹, BbªJ×AíR, % }F)`íälN™‰b, úk«É−‘íã ¿^‹íüæÊÝ(4íÔH, ˇé|f$(4_%%.?šƒ’eí.ú˚4í!Z, ÄÑ; W…dF)ƒíÝ(4Ä‹É[ì!‹, ÊìälN™úõ”%ÈAÅ0íÝ(4ã¿?‰j Þ, zp7Ý(4Ô”íæÊuóçËpéí9õ FJ, _Ë.JÝ(4íi }&’eF¿ Rím7, @çu_~7/.bí. 5 !D ‡ ¬ , Ì%Èç6C\œÉ4BkøO×V, %‚à% }F)`í−älN™, JàV ã¿«É„Ví−², Í7®N™‰bu´öxíälkõ”%ÈAÅ0íN™4<°, ºÿ ˇÃãí$lj¶íð„ 7¬ d.ÊÄ‹É[ínv, ¢ókl¾j¶íÌ„, Ĥ×ÖJ( 4É[T«n, OrÖóÉd.ºN|−šx.ú˚CÝ(4íÔ4, ¥6<‚O, mÿ(4 _n‰bÈíÄ‹É[, ª?}³Õ¥óçÖí’em7, 7)ƒRÏíì!‹ F Hiemstra and Jones (1994) íÝ(4Ä‹É[ìíT|, ‰bÈÝ(4¶}ím7?\yüí¾¦ |V, 7BbøFíõ„!‹)k[ 7, JjZè6©è %â[ 7 õ„!‹cÜíêÛ, Ê(4Ä‹É[í¶}, "×ÖbíälN™‰b, îb²äl kõ”%ÈAÅ0, ²k5, % }F)`í−älN™üõxeäl−²í?‰, ÝBÊ« ÉË–ê ÎÞ £½É|¨M‰0áñ, ?§õ”%ÈAÅ0í à, 7`¨“hQ¿ÀN b‰0¸«ÉË–ê ÎÞ yDõ”%ÈAÅ0È, ×Ûó àí¥—É[, ¥ø%ÈíA. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(25) 19 Åúkõ”Þ?x#Ëí:4 ⤪c, älN™Î7ªJ¥ø−í‰, ?D%ÈAÅ O#ËíÉ:4 Ĥ, Ê(4Ä‹É[íì2, Î7älN™üõ.ÜÑã¿«É−ší½ b¡55Õ, ?xÌÀ¢ÞCuõ”ÞÌ¿§−ší à Ou, ʽÉ|¨M‰0jÞ, w(4Ä‹É[.éO, u´H[¤‰b._¯TÑälN™? ÊBbb-sií!í5‡, lhô[ 7 íÝ(4Ä‹É[ì, êÛ½É|¨MD«ÉË– ê ÎÞ ‰0úõ”%ÈAÅ0Ý(4íäl^‹, ½É|¨Múõ”%ÈAÅ0ÖÍ. x(4Ä‹É[, OuºÝ(4älkõ”%ÈAÅ0, ªc(4_F.?|ƒm7, Ý(4 _ºªAŠíšƒwÝ(4Ä‹É[Ô4 Ĥ, âÝ(4Ä‹É[íì, ½É|¨M?ª TÑã¿«É−ší¡5, °v6?TÑ¥ø−‰íGßälN™ 7Ê«ÉË–ê Î Þ ‰0úõ”%ÈAÅ0íÝ(4Ä‹É[ì¶}, êÛv‰bD½É|¨M?xó°í Ý(4¥—^‹, ¤s_älN™D%ÈAÅ0ÈOÝ(4íÉ:4, 7¥¶}úÎ(4Ä‹É[ ì, êÛ¤s‰bD%ÈAÅ0Èí¥—^‹, ŸlÊ(4Ä‹É[ì2, 1„\ì|V, Ou %âÝ(4 Granger Ä‹É[ì(, ’eíÝ(4Ô4?ªâì7\Bbšƒ …d´rÖËjªJ‹JZª, Wà…dÉSàW\Í%È qãº}F)`í −²N ™ 2íã¯älN™, TÑÄ‹É[ìí«n, Í7, u´wFH‰b?y_~í¥ø− íš? 1/yx%È<2 wŸ, ÊFÅ4íóɇæn, …Åíõ”%ÈAÅ0, u´§ƒ× ÅíälN™‰bí à? ¥<·uªJ./Økû˝í‡æ. ¡5d. [1] Š× (2003), «ÉD1ÅsË−=N™É:45û˝, Å AŠ×çò¼Üû˝Fî =d [2] Š²¸ôo (1993), «É°vDälN™í,lDwì: 1968∼1991, %Èd…˚, 21, 123–159. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(26) 20 [3] Š² ô²Ë 2” (1998), −=~õwìD%ÈAÅ0ã¿, %Èd…˚, 26, 431–457 [4] =¤¸2” (2001),. ÉH«Éí−=: ïªÚ²_D³g‚š¶í@à, A. d£þ}çÕ…˚, 13, 515–540 [5] Àq ò/† 2” (2002), 1ÅDn…úÅqß|5 à}&, 2002 ,ñ%Èl¾_ ûn} [6] ˜_ (2002), À¢‰bD%ÈAÅ5É[, .C×çÀ¢Fî=d [7] ÏDQ 2M (2003), À¢älN™Dõ”älN™m7ø_ý? «ÉälN™íõ„}&, Ad£þ}çÕ…˚, 15, 627–660 [8] Aoki, M. (1992), Interactions of Real GNP Business Cycles in a Three Country Time Series model, Working paper No.675, University of California, Los Angeles. [9] Baek, E. and W. Brock (1992), A General Test for Nonlinear Granger Causality: Bivariate Modle, Working Paper, Iowa State University and University of WisconsinMadison. [10] Brock, W. A., W. D. Dechert, J. A. Scheinkman,and B. Lebaron (1996), A Test for Indenpendence Based on the Correlation Dimension, Econometric Reviews, 15, 197– 235. [11] Brock, W. A., Hsieh, D. A., and LeBaron, B. (1991), In Non-linear Dynamics, Chaos and In Non-linear Dynamics, Chaos and Instability: Statistical Theory and Economic Evidence, MIT Press, Cambridge, MA. [12] Camacho, M. and G. Perez-Quiros (2002), This is What the Leading Indicators Lead, Journal of Applied Econometrics, 17, 61–80. [13] Chauvet, M. (1998), An Econometric Characterization of Business Cycles Dynam-. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(27) 21 ics with Factor Structure and Regime Switching, International Economic Review, 39, 969–996. [14] Chen, S. W. (2001), A Note on Taiwan’s Business Chronologies in Terms of the Markov- Switching Model, Taiwan Economic Review, 29, 153–176. [15] Chen, S. W. and J. L. Lin (2000a), Modeling Business Cycles in Taiwan with TimeVarying Markov-Switching Models, Academia Economic Papers, 28, 17–42. [16] Chen, S. W. and J. L. Lin (2000b), Identifying Turning Point and Business Cycles in Taiwan: A Multivariate Dynamic Markov-Switching Factor Model Approach, Academia Economic Papers, 28, 289–321. [17] Copeland, T. E. (1976), A Model of Asset Trading under the Assumption of Sequential Information Arrival, Journal of Finance, 31, 1149–1168. [18] de Brouwer, G. and J. Romalis (1996), External Influences in Output: An Industry Analysis, Research Discussion Paper 9612, Economic Research Department, Reserve Bank of Australia. [19] Dickey, D. A. and W. A. Fuller (1979), Distribution of the Estimators for Autoregressive Time Series with a Unit Root, Journal of American Statistics Association, 74, 427–431. [20] Dickey, D. A. and W. A. Fuller (1981), Likelihood Ratio Tests for Autoregressive Time Series with a Unit Root, Journal of American Statistics Association, 49, 1057–1072. [21] Durland, J. M. and T. H . McCurdy (1994), Duration-Dependent Transitions in a Markov Model of U.S. GNP Growth, Journal of Business and Economic Statistics, 12, 279–288. [22] Engle, R. F. and C. W. J. Granger (1987), Co-integration and Error Correction :. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(28) 22 Representation, Estimation, and Testing, Econometrica, 55, 251–276. [23] Filardo, A. J. (1994), Business-cycle phrases and their transitional dynamics, Journal of Business and Economic Statistics, 12, 299–308. [24] Ghysels, E. (1994), On The Periodic Structure of the Business Cycle MarkovSwitching Model, Journal of Business and Economic Statistics, 12, 289–298. [25] Hamilton, J. D. (1989), A New Approach to the Economic Analysis of Nonstationary Time Series and Business Cycle, Econometrica, 57, 357–384. [26] Hamilton, J. D. and G. Perez-Quiros (1996), What do Leading Indicators Lead? Journal of Business, 69, 27–49. [27] Hayo B. (1998), Money-Output Granger Causality Revisted: An Empirical Analysis of EU Countries, Provided by the Economics Department of Washington University. [28] Hiemstra, C. and J. D. Jones (1994), Testing for Linear and Nonlinear Granger Causality in the Stock Price-Volume Relation, Journal of Finance, 49, 1639–1665. [29] Hiemstra, C. and C. F. Kramer (1993), Accounting for Stock-Return Dynamics with a Macrofactor APT/factor-ARCH Model, Unpublished manuscript. Glasgow, Scotland: University of Strathclyde, Department of Accounting and Finance, and Washington, DC: International Monetary Fund, Research Department. [30] Hsieh, D. (1991), Chaos and Nonlinear Dynamics: Application to Financial Markets, Journalof Finance, 46, 1839–1877. [31] Huang, C. H. (1999), Phases and Characteristics of Taiwan Business Cycles: A Markov- Switching Analysis, Taiwan Economic Review, 27, 185–213. [32] Ioannis, A., D. Ayling, W. M. Mahmood (2000), Nonlinear Granger Causality in. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(29) 23 the Currency Futures Returns, Economics Letters, 68, 25–30. [33] Jaditz, T. and J. Jones (1992), Granger Causality Between the Consumer and Wholesale Price Index, Unpublished Manuscript, Washington, DC: Bureau of Labor Statistics and Securities and Exchange Commission. [34] Johansen, S. (1991), Estimation and Hypothesis Testing Cointegration Vectors in Gaussian Vector Autogressive Model, Econometrica, 59, 1551–1581. [35] Kim, M. J. and J. S. Yoo (1995), New Index of Coincident Indicators: A Multivariate Markov-Switching Factor Model Approach, Journal of Monetary Economics, 36, 607– 630. [36] Kim, C. J. and C. Murray (2002), Permanent and Transitory Components of Recessions, Empirical Economics, 27, 163–183. [37] Koˇcenda, E. (2003), An Alternative to the BDS Test: Integration Across the Correlation Integral, Econometric Reviews, 20, 337–351. [38] Kim, C. J. and C. R. Nelson (1998), Business Cycle Turning Points, A New Coincident Index, and Test of Duration Dependence Based on a Dynamic Factor Model with Regime Switching, The Review of Economics and Statistics, 80, 188–201. [39] Koˇcenda, E. (2001), An Alternative to the BDS Test: Integration Across the Correlation Integral. Econometric Reviews, 20, 337–351. [40] Kunfong and Yong-Yil (1997), Business Cycle in Korea Is There Any Stylized Feature ?, Journal of Economics Studies, 24, 275–293 [41] Lahiri, K. and J. G. Wang (1994), Predicting Cyclical Turning Points with Leading Index in a Markov-Switching Model, Journal of Forecasting, 13, 245–263. [42] Layton, A. P. (1996), Dating and Predicting Phase Changes in the U.S. Business. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(30) 24 cycles, International Journal of Forecasting, 12, 417–428. [43] Layton, A. P. (1998), A Further Test of the Influence of Leading Indicators on the Probability of US Business Cycle Phase Shifts, International Journal of Forecasting, 14, 63–70. [44] Lucas, R. E. (1977), Understanding Business Cycle, in K. Brunner and A. Meltzer (eds.), Stablilization of the Domestic and International Economy, Carnegie-Rochester Series on Public Police 5, 7–29. [45] McGough T. and S. Tsolacos (1995a),Property Cycles in the UK: An Empirical Investigation of the Stylized Facts, Journal of Property Finance, 6, 45–62. [46] Pesaran, M. H. and Y. Shin (1998), Generalized Impulse Response Analysis in Linear Multivariate Model, Economics Letters, 58, 17–29. [47] Scheinkman, J. A. and B. LeBaron (1989), Non-linear Dynamics and Stock Returns, Journal of Business, 62, 311–337. [48] Stock, J. H. and M. W. Watson (1989), New Indexes of Coincident and Leading Economic Indicators, in O. Blanchard and S. Fischer (eds), NBER Macroeconomics Annual, Cambridge, MA: MIT press. [49] Stock, J. H. and M. W. Watson (1991), A Probability Model of the Cointcident Economic Indicators, in K. Lahiri and G. H. Moore (eds.), Leading Economics Indicatos: New Approach and Forcasting Records, Cambridge, Cambridge University Press.. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(31) 25. [ 1: ®älN™‰bíš…’ecÜ. áñ. !‚. ÀP. ÇávÈ. 1996 . —. 1975 ø. `¨“©A©~Ì v. —. üv/~. 1962 ø. ½É|¨M. —. ìNj. 1960 ø. ŒÞX# M1b. —. ìNj. 1982 ø. R»ÓgNb. 2001 . —. 1960 ø. $gNb. 1966 . —. 1967 ø. `¨“hQ¿ÀNb. «ÉË–ê ÎÞ. —. —jt. ’eVÄâW\Í%È qãº}TX, …û˝c. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics. 1981 ø.
(32) 26. [ 2: ®älN™‰bíÀ;ì!‹. ™Ìcq: À;. t ì$l¾. ]ô‚b. ì!‹. `¨“hQ¿ÀNb. −3.034. 5. I(1). `¨“©A©~Ì v. −2.979. 17. I(1). ½É|¨M. −0.621. 16. I(1). ŒÞX# M1b. −1.570. 5. I(1). R»ÓgNb. −0.933. 1. I(1). $gNb. −1.515. 2. I(1). «ÉË–ê ÎÞ. −0.243. 7. I(1). ‰b. (1) À;ì t $l¾í 5% éO®Ä@äMÑ −3.60 (2) ]ô‚bÑ¡Î Ljung-Box ì{ÏåóÉF²¦5|_]ô‚b. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
(33) 27. [ 3: älN™Dõ”%ÈAÅ0Èuc¯É[ì!‹ ™Ìcq: Ìuc¯É[. Zt ì$l¾. uc¯É[. −2.572. .æÊ. −10.702*. æÊ. ½É|¨M. −1.975. .æÊ. ŒÞX# M1b. −3.134. .æÊ. R»ÓgNb. −1.134. .æÊ. $gNb. −2.397. .æÊ. «ÉË–ê ÎÞ. −1.330. .æÊ. ‰b. `¨“hQ¿ÀNb. `¨“©A©~Ì v. (1)Zt Ñ Philips-Ouliaris t ì$l¾ (2)****** }[ýÊ 1% 5% 10% í®Ä-éO. 第五屆全國實證經濟學論文研討會 The 5th Annual Conference of Taiwan's Economic Empirics.
相關文件
Students are asked to collect information (including materials from books, pamphlet from Environmental Protection Department...etc.) of the possible effects of pollution on our
which can be used (i) to test specific assumptions about the distribution of speed and accuracy in a population of test takers and (ii) to iteratively build a structural
本研究採用三種判斷準則來比較 Nelson-Siegel Model、Extend Nelson-Siegel Model 與 Nelson-Siegel-Svensson Model 的配適能力，配適結果如表 4 表示，其中
另外關於前現代的 西藏醫學發展與佛教關係的探討，參考 Gyatso (2015), Being human in a Buddhist world: an intellectual history of medicine in early modern Tibet,
文學賞析能力：形式、節奏的感 悟、從詩歌與政治的關係了解詩 歌的怨刺作用，探討深層的主題
Standard approach to pricing CDOs is the one-factor Gaussian copula model with constant pairwise correlations, constant CDS spreads, and constant default intensities for all
二、 本計畫已將部分研究結果整理，發表於國際研討會(Chan, Y.-H., Lin, S.-P., (2010/7), A new model for service improvement design, The 2010 International Conference
D.Wilcox, “A hidden Markov model framework for video segmentation using audio and image features,” in Proceedings of the 1998 IEEE Internation Conference on Acoustics, Speech,
