: 3 ( 510632) ( ) :,,,,,, Mahalanobis,, : : ;, (,2002), - (Dow Jones Sustainability Index),, (2001Π394ΠS ),, 2003 1300, 50 %,,, 1 1990 2003 ST, 2004 3 1,1283 200, 514 1 2000 257,2000 1026, 3 2004 4 2004 6 (IACMR),,, (031909) ST,, ST,, (000533) 2002 4 23 ST,, 2004 3 25 ST (600629) (600633) ST (600763) ST (000585) 26 (2004) 106
2005 1 194 ( 413 1), (2003) 1991 2001,,,, 1 1990 2003 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 6 6 39 123 111 24 207 206 104 100 136 84 71 66 1283 ST 3 1 3 17 30 33 9 58 25 11 7 4 2 200 3 1998 3 16 ST ( :CSMAR ),,,,(1998) (1999) (2001) (2002) (2003) ;,, (2000) (2000) (2000) (2001) (2003) :, ;, ;, (power of test),,,,,,,,,,, (panel data),,,,,, Mahalanobis,, :;,;, ;, :, (2000) (ROE),,, 2000 257 6 ST, 4218 1, 107
: ; (1998) (2000), ( EPS) ;(2000),,,,,,,,,, 50 ;, 50 ; (2000) (2000),,,,, (1999) 27 ST 27 ST,, ST ; (2001) 70 ST 70 ST,, (Logit model), (2002) 42 ST 42 ST, Logistic,,,,,,, t -,,, (2001), ; (2002) Higgins Horne, 1994 282, 21 ;(2003),, 21,,,,, 157 4,, ; 158,,,,,, 108
2005 1,,,,,,,,, : Y i, t = ( y 1, i, t,, y m, i, t ) i t, m, i = 1,, N, t = 1,, T Y i, t, (Mean reversion),y i, t,, Y i, t (Lag) p (VAR( p) process) : Y i, t = a + b, k + Y i, t - 1 1 + + Y i, t - p p + k, t (1), a = ( 1, a,, m, a ), b, k = ( 1, b, k,, m, b, k ) ( ST ) k a ( k = 1,, T), k = 1, b,1 = b, ST 1,, p m m VAR, p Akaike (AIC) Schwartz (BSIC) i, t m (i1i1d1), E ( i, t ) = 0, - 6 = p E ( i, t i, t ), E ( i, t j, s ) = 0, Π i j, t s (1) : l = - N T ln2-2 1 2 ln 6-1 p 2 i, t 6-1 p i, t (2) (2) ^ a, ^ b,^ 1,,^ p - 6^ p, AIC p (AIC = ln 6 p + 2 mπn T, Judge, et al. 1985, p. 870),, Y i, t Bar2Hen and Daudin (1995) Bedrick, et al. (2000), ST Mahalanobis (Mahalanobis generalized distance) : d = b 6-1 p b (3) Mahalanobis,, i t Z i, t ( Y i, t,y i, t - 1,,Y i, t - p ) :, Z i, t = 0 + (Y i, t - a - Y i, t - 1 1 - - Y i, t - p p ) 1 = 0 + b 1 + i, t 1 (4) Y i, t 1993, VAR 1994, k = 1,,9, k > 1, ST,, k = 1 b b,1 109
: 0 = ( b 6-1 p b )Π2 d = dπ2 (5) 1 = - ( 6-1 b)πd (6) p, 0 1,Mahalanobis, ST Z i, t - dπ2 dπ2, 1 :,, S i, t = min( S i, t - 1 + Z i, t -,0) < - > 0, > 0 (7), S i, t i t, S i,0 = 0 (sensitivity parameter) Z i, t, S i, t = 0 Π t ST,, S i, t < -, Z i, t i t - 1, S i, t, ST, ST,: E(C) = f f (, ) + s s (, ) (8) min,,, f s, f (, ) = Pr ( S i, t > - ST ) ST (), s (, ) = Pr ( S i, t - ) ST (), (jackknife) :,, f = s = 015 ;, (uniformly distributed) [0,015 d ] [0,5 d ], 30 30 = 900 (, ) ;, ST, (2) VAR, (4) Z i, t ;, (7) S i, t,;,(, ), 250, ST f ST s ;, 900,(, ) f s ;, (8), f s, (, ) ;,,,, (7), 1996 516 ( 151 ST ) 1993 2002, 1997 206 ( 25 ST ) 1994 2002 CSMAR S i, t ( Neftci,1985 ;Chu and White,1992),,S i, t, 110
2005 1 (,2000),, 21, 28, 467 ( 132 ST ),, 15 ( 2) : (1) ; (2) ; (3) ; (4) ; (5) ; (6) 2 Π Π Π Π Π Π Π ( + )Π ( + )Π Π ( - )Π ( - )Π ( - )Π ( - )Π (2002) (2003) 1998,10 %,,10 %, 10 % 11 % 200,9 % 10 % 20,2001 6 %, 6 % 21 6, SGR =,A, S m (1 - d) A E A S - m (1 - d) A E,, m, d, E 111
:,467 9,, (panel data unit root test) : (1) Levin2Lin2Chu :, (Levin, et al. 2002) (2) Im2Peraran2Shin gt - :, ( Im, et al. 1997) (3) Im2Peraran2Shin LM:, ( Im, et al. 1997) 3, 5 %,1 %,,, 3 Levin2Lin2Chu Im2Peraran2Shin gt - Im2Peraran2Shin LM - 38160 3 3-3122 3 3 3193 3 3-51108 3 3-4118 3 3 5111 3 3-43139 3 3-3165 3 3 4135 3 3-34190 3 3-2140 3 3129 3 3-54122 3 3-4134 3 3 3185 3 3-38173 3 3-3126 3 3 2179 3-53158 3 3-4137 3 3 3181 3 3-46101 3 3-3166 3 3 3138 3 3-31140 3 3-3105 3 3 3106 3 3-24111 3 3-2161 3 3 2175 3-36128 3 3-3170 3 3 4118 3 3-33173 3 3-3147 3 3 3192 3 3-40108 3 3-3159 3 3 4108 3 3-24135 3-2168 3 3 3126 3 3-38120 3 3-3140 3 3 3177 3 3 N > T = 10 = 1 % - 24141-2142 2181 = 5 % - 23194-2134 2169 = 10 % - 23169-2130 2162 : 3 3 3 t - 1 % 5 % Levin, Lin and Chu (2002) Im, Peraran and Shin(1997),, GAUSS BHHH VAR, ( m = 6, ) 144,,AIC, 1, k > 1 ( b b,1 b,2 ),AIC VAR 1, 0, VAR, 112
2005 1 VAR, (t - ) : a = 0. 6855 0. 0152-0. 4006 0. 0083-0. 1417-0. 8225 (8. 295) (1. 140) ( - 3. 191) (0. 903) ( - 2. 885) ( - 12. 307) ST b = 0. 0019 0. 0025-0. 0613 0. 0626-0. 1217 (3. 183) (4. 259) ( - 9. 431) (10. 950) ( - 12. 628) VAR 1 = 0. 1855 0. 2117 0-0. 0235 0. 1180 (5. 861) (4. 060) (0) ( - 2. 904) (3. 622) 0 0. 6281 0. 5075 0 0 (0) (4. 009) (7. 144) (0) (0) 0. 1482 0 0. 3382-0. 481 0. 1641 (3. 091) (0) (3. 855) ( - 6. 309) (4. 281) - 0. 1052 0 0 0. 5083-0. 318 ( - 4. 828) (0) (0) (9. 140) ( - 3. 645) 0. 4205 0. 394 0. 097 0 0 (7. 929) (4. 388) (2. 863) (0) (0) 0. 2671 0. 0859 0. 1006-0. 875 0. 2101 (3. 157) (2. 850) (3. 544) ( - 6. 592) (3. 593) 6 p = 0. 0284 0. 0266-0. 1519 0. 155-0. 0853 0. 0266 0. 1894-0. 1906 0. 1722-0. 0682-0. 1519-0. 1906 0. 1388 0. 0862 0. 1039 0. 155 0. 1722 0. 0862 0. 1451-0. 205-0. 0853-0. 0682 0. 1039-0. 205 0. 0089 0. 1724 0. 1839 0. 0647-0. 1363 0. 0147 ^ 1 :,, ; ; ; ; ;,, (3) (5) (6) ST Mahalanobis ^ 0 ^ 1 : d = ^ b - 1 p ^ b = 017012 ^ 0 = ( b 6-1 p ^ b )Π2 d = dπ2 = 013506 113
: 1 = = - ( 6-1 p b )Πd = [1. 1771 0. 2148 1. 3654-0. 5689 0. 5259 1. 7320] 1 ST 1, Z i, t, (4) Z i, t 1 335 132 ST Z i, t ST Z i, t dπ2 = 013506 - dπ2 = - 013506, Z i, t, Z i, t, ST Z i, t 1997 ; ST,ST Z i, t - dπ2,,,,,, 900 (, ),, (, ) 4 f s, f s, f E(C) s f (7), 013 010628 013811 010912 010482 011916 4 0135 010835 01227 010984 010355 012152, f = 0155 s = 0145 014 011146 110824 011117 010626 011853,,, = 011304 = 017184,ST f = 011758, ST s = 010421, : S i, t = min ( S i, t - 1 + Z i, t - 011304,0) < - 017184 114 (9) 0145 011849 018577 011416 010569 012451 015 010597 016208 011396 010733 012059 0155 011304 017184 011156 010421 011758 016 010943 016927 011749 010824 012366 0165 011826 110545 011705 010902 012138 017 010737 018268 011712 010495 012233
2005 1, Z i, t < 011304,S i, t 0 S i, t < - 017184,,(9) 1997 206, 29,21 ST,ST 25 4, f = 0116, s = 010442,,, 17 ST - 017184,, f,st ST ST 9 ST (),ST,, :, ; ( ST) Mahalanobis,,, ST, ;,, ;,,,,ST,2003 :, 5,1999 :, 4,2002 :, 3,2002 :, 12,2002 :,3,2000 :, 3,2003 :,9,1998 :,10,1999 :, 7,2003 :, 1,2004 :, 8,2000 :, 1,2000 :,1,2001 :, 6 115
:,2001 :, 1,2000,, 9,2000,, 4,1998 :, 7,2001,,5,2000 :, 6 Bar2Hen, A1 and J1J1 Daudin, 1995, Generalization of the Mahalanobis distance in the mixed case, Journal of Multivariate Analysis 53, 332 342. Bedrick, E.J., J. Lapidus and J. F. Powell, J. F., 2000, Estimating the Mahalanobis distance from mixed continuous and discrete data, Biometrika 56, 394 401. Sons. Chu, C.J. and H. White, 1992, A direct test for changing trend, Journal of Business and Economics Statistics 10, 289 299. Im, K., M. Pesaran and Y. Shin, 1997. Testing for unit roots in heterogeneous panels. University of Cambridge Working Paper. Judge, G., W. Griffiths, R. Hill, H. Lutkepohl and T. Lee, 1985, The Theory and Practice of Econometrics. New York : John Wiley and Levin, A., C. F. Lin and C. S. Chu, 2002, Unit root tests in panel data : Asymptotic and finite sample properties. Journal of Econometrics 108, 1 24. Neftci, S. N., 1985, A note on the use of local maxima to predict turning points in related series. Journal of the American Statistical Association 80, 553 557. The Sustainable Development of China s Listed Firms : Econometric Model and Empirical Evidence Su Dongwei (College of Economics, Jinan University) Wu Yangru (Andromeda School of Business, Rutgers University) Abstract :Sustainable development of listed firms builds a solid foundation for the sustainability of securities market. As China s central government pays more and more attention to the development of securities market, it is very important to measure the competitive edge, long2term performance and growth prospects of the listed firms. This paper uses a novel approach in developing such an econometric model. It also uses, for the first time, panel data econometric methods in conducting empirical investigations on China s listed firms. By using state2of2the2art statistical methodologies including Mahalanobis distance, computer intensive method and Jackknife, the paper explores the determinants of long2term performance of listed firms and finds that a dynamic cumulative performance model can successfully capture the intriguing relationship among various proxies for sustainability, disclose the intrinsic characteristics of sustainable firms, and predict the future performance for most of the listed companies. Key Words :Listed Firms ; Sustainable Development ; Dynamic Cumulative Performance Model ; Panel Data Econometrics ; Chinese Securities Markets JEL Classification : G300, G150,C330 ( : ) ( : ) 116