( 413 1), (2003) ,,,,

Size: px

Start display at page:

Download "( 413 1), (2003) ,,,,"

琏峻解
7 years ago
Views:

1 : 3 ( ) ( ) :,,,,,, Mahalanobis,, : : ;, (,2002), - (Dow Jones Sustainability Index),, (2001Π394ΠS ),, , 50 %,,, ST, , , , , (IACMR),,, (031909) ST,, ST,, (000533) ST,, ST (600629) (600633) ST (600763) ST (000585) 26 (2004) 106

2 ( 413 1), (2003) ,,,, ST ST ( :CSMAR ),,,,(1998) (1999) (2001) (2002) (2003) ;,, (2000) (2000) (2000) (2001) (2003) :, ;, ;, (power of test),,,,,,,,,,, (panel data),,,,,, Mahalanobis,, :;,;, ;, :, (2000) (ROE),,, ST, , 107

3 : ; (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, , 21 ;(2003),, 21,,,,, 157 4,, ; 158,,,,,, 108

4 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 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 ln 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 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

5 : 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 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 ], = 900 (, ) ;, ST, (2) VAR, (4) Z i, t ;, (7) S i, t,;,(, ), 250, ST f ST s ;, 900,(, ) f s ;, (8), f s, (, ) ;,,,, (7), ( 151 ST ) , ( 25 ST ) CSMAR S i, t ( Neftci,1985 ;Chu and White,1992),,S i, t, 110

6 (,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, %, 6 % 21 6, SGR =,A, S m (1 - d) A E A S - m (1 - d) A E,, m, d, E 111

7 :,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 N > T = 10 = 1 % = 5 % = 10 % : 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

8 VAR, (t - ) : a = (8. 295) (1. 140) ( ) (0. 903) ( ) ( ) ST b = (3. 183) (4. 259) ( ) ( ) ( ) VAR 1 = (5. 861) (4. 060) (0) ( ) (3. 622) (0) (4. 009) (7. 144) (0) (0) (3. 091) (0) (3. 855) ( ) (4. 281) ( ) (0) (0) (9. 140) ( ) (7. 929) (4. 388) (2. 863) (0) (0) (3. 157) (2. 850) (3. 544) ( ) (3. 593) 6 p = ^ 1 :,, ; ; ; ; ;,, (3) (5) (6) ST Mahalanobis ^ 0 ^ 1 : d = ^ b - 1 p ^ b = ^ 0 = ( b 6-1 p ^ b )Π2 d = dπ2 =

9 : 1 = = - ( 6-1 p b )Πd = [ ] 1 ST 1, Z i, t, (4) Z i, t ST Z i, t ST Z i, t dπ2 = dπ2 = , 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), , f = 0155 s = ,,, = = ,ST f = , ST s = , : S i, t = min ( S i, t Z i, t ,0) < (9)

10 2005 1, Z i, t < ,S i, t 0 S i, t < ,,(9) , 29,21 ST,ST 25 4, f = 0116, s = ,,, 17 ST ,, 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

11 :,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, Bedrick, E.J., J. Lapidus and J. F. Powell, J. F., 2000, Estimating the Mahalanobis distance from mixed continuous and discrete data, Biometrika 56, Sons. Chu, C.J. and H. White, 1992, A direct test for changing trend, Journal of Business and Economics Statistics 10, Im, K., M. Pesaran and Y. Shin, 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, 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, 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

Microsoft Word - A201202-493_1329751213.doc

Microsoft Word - A201202-493_1329751213.doc 5 10 15 20 25 BP 神经网络在中国创业板企业成长性预测研究 ** 孙静稳, 刘金平 ( 中国矿业大学管理学院, 江苏徐州 221116) 摘要 : 根据创业板企业的高科技和高成长性特点, 成为金融证券市场热门关注的对象, 其成长性研究是资本