4 5 6 7 8 9 : < = > ? @ A B C D E
?
F > G HFTSE
7 8 9 : I J < K L M N O P Q R S T U V W X T Y Z \∗
O P S T W X T Y ! " #:
$ % & ' ( ) * + , - . * / 0 1 2JEL
3 4 5 6:
G30, M14
∗ ] ^ _ ` :a b ,c d e f g h i j h k ,l m n 1163 o p q r s t u 64v w x y : 0920671 950;z { : (02) 2939-8004; E-mail: 92352506@nccu.edu.tww | } ~ @ [ \ ] ^ _ ` { | } ~ q ] w 3 ~ , _ ` w
corporate social responsibility
¡ ¢ £ ¤ ¥ ¦ § ¨ © ª « ¬ ® ¯ ° ± ² ³ § ¨ ´ µ ¶ · ¸,
¹ º » ¼ ½ ¾ ¿ À Á ´ Â Ã ³ Ä Å,
Æ Ç È É ³ Ê Ë Ì ¼ Í Î Ï Ð Ñ Ò Ó Ô Õ ³ Ö × Ó Ø Ù Ú,
Û Ï selection bias
³ © FTSE All-share
Õ ¢FTSE
Õ ³ ,
º ² ,
Rubin
1973
Rosenbaum and Rubin
1983, 1985a, 1985b
Ø ³ µ matching method
,
Ò Ó Ô Õ ´ ! " # $ Ï µ,
# % È É µ & ³ Í Î Ï ³ ¥ ¦ § ¨ Ö ×,
' · ª ³ © « ¬ Ê Ë ( ) ¼ ³ * + § ¨ , ¢ ² ³ ¬;
´ - .,
/ È ³ ¬ 0 1 ³ ¥ ¦ § ¨ ´ µ ° ·,
Æ Ç 2 3 ³ È É Ê Ë 4 5 6 7 8 9 : ; < =,
$ µ ¥ ¦ § ¨ ³ > ? ¨ Ë @ ©1.
A B $ % C D E F G H I J K L M N O,
P Q R S T U V W X Y Z [ * \ ] ^ _ ` a b S c d e ] & f g h i j k M ( ),
1 l m n o p q r s t u v w x t y z { | } ~ x n n 2) *,
o p q r s t + , -. / 0 1 2 3,
4 5 / s + , - 6 7 8,
9 : < l = > ? @ A B C D E F G H x I * J K L M N O n P Q R S T,
U V W X 2 3 n o p q r s t Y Z \ ] ^ _ ` a b N p n c d 3 ! } ",
n # 8 $ % & ' ( 7 8 ) * n $ % + ,,
- . / 0 1 2 3 J 4 5 6 7 F 8 9 o : n l ; < p Y J / 0 = > ? R 5;
@ * n [ \ J 0 ] + , Y - ^ G _ ` 7 e f g \ h n / 0,
c i Y j k d l n H = < m f n o;
& 3 m 9 p | - E f q ) D E r s t u l m v w,
x y,
B C l = J z { K L | } s ~ s Z ,
Z + F B C l = F < o p q r s t s ~ p 2 3 _ ?
~ } u Z ,
7 d Fri-edman
1970
,
Z x Y D E B C ( a b 0 M N O l n <;
Brag-don and Marlin
1972
t o p q r Y g x I F p > \ HF j k ?
,
.Vance
1975
FAupperle et al.
1985
JUllmann
1
l m n o p q r s t u world business council for sustainability and development v w n o x u :y n o z { | q } ~ , r s ¡ ¢ £ ¤ ,¥ ¦ § ¨ \ © ª « ¬ ® ¯ ° ± ² x p ³ ª x u ~ ´ µ ¶ · ¸ ;¹ º ~ v w » ¼ ½ Frooman 1997 ® Carroll 1979 ] McWilliams and Siegel 2001 w
2
c ¾ ¿ { ~ n o À Á Nike® Adidas® SonyÂ Ã Ä Å « Æ Ç È É Ê ,v Ë Ì Í « Î Ï Ð Ñ ® Ò Ó Ô Õ Â Ö × ;Ø Ù ,{ Ú Û Nokia Ü | y Ý Þ × ß ~ à á ,â ã u ä ª å æ ~ ç è é Å à á ~ Ý Þ ¸ ~ ê ë ì ,« í ¶ î ï ð ñ ò Ý Þ ó ô ª õ ö ~ à á Ù ,É ÷ ø í ¶ Å Ò Ó ® ù ú ] û ú í ´ ~ ü ý ,¥ þ ÿ ù ú ;] Unilever ] l m Ô Õ ´ Ï Ú i u WWF _ d è p q , 2005 ¡ 60%~ ¢ £ ¤ ¥ ¦ ;c § ,¨ © ª ¨ « s ¬ l ® ¯ o p ° ± ² n o ³ ! ," # $ % ~ & ' ( s w 3 è e f Ë g h ¯ ¡ { ~ i À j , 1929 k l m n o | ¡ p q n o 10%~ õ r ,« Ä Å p q s t ´ í & ' ¬ u n o ë v « è j w x ¬ ± ~ y ( z { ,« t j w | © } ~ Ö ¯ ° × o â ã r Ø õ â Ù Ú Û w Û Ü Ý p q Þ ß ,« Ü Ý à á â :y $ ã î d ~ × ~ ä å × j õ ö ~ % , ê æ ~ ç è é ³ ê ë ì Ö × ~ í î w Á ï ð ñ Ö × ~ ,ò ó ¡ ô õ ~ ö ÷ ,ø ò ó ã û ù ú ñ Ö Ú × ~ ,ã û û á « ü × ~ â ÷ ø õ ö % ,ý þ ÿ r Ø õ w ¸
1985
r 7 R c " V;
l, Walley and Whitehead
1994
JHenderson
2002
t o p N l = > ? n ! " | &,
v w 3 # f [ \ $ { % & n " ' i { (;
w }, Becchetti et al.
2007
) * H + u / 0 J o : 5 , F + , - . J l m / 0 ] t N 1 + 2 3 l m $ % 4 5 n 6 K L M N O s t,
x y < 7 8 9 x I n o p q r s t Y Z K $ % 2 3,
: Y y 5 ; t 4 5 2 3 < = shift of focus hypothesis
> !
,
t ? | } o p q r,
- .Bowen
1953
JArrow
1973
t > ? J @ A o p x / B B R 5,
x y C } D E F & KG 5 , H
stakeholders
n I =,
J ( o p * | ( o p;
c i, Moskowitz
1972
FParket and Eibert
1975
JSoloman and Hansen
1985
0 t K o p q r n 3 = N ^ L x I * l m 2 3 & M N O n " V l
,
t P / 0 Q @ * n 0 ] + , Y & R ( = > ? n t g Turban and
Green-ing, 1997
,
< S T R l = s t g v w J U ' Bowman and Haire,1975;
Alexander and Bucholtz, 1978
8 9 \ V W X J ? * p x I j k Porter
and van der Linde, 1995; Fombrun et al., 2000
,
c i Y \ Z B D < [ n \ ]
Spicer, 1978; Moussavi and Evans, 1986
,
^ ~ j k _ ! ` a < b n H = < m ~ f 8 9 _ ! < b H = i n - ? Tsoutsoura, 2004
,
W c t l m n \ V W X J 0 1 2 3 c d - e Werther and Chandler, 2005; Peloza, 2006
Cornell
and Shapiro
1987
JPreston and O’Bannon
1997
t f - g ^ } Co p h i j K G 5 , H n k l
,
Y p l m n 0 1 2 3 & M n R =,
m g n o p 2 3 n + Y m o n 0 1 2 3,
{ p u M n 5 ,,
: Y y q 5 ; t o p S r < = social impact hypothesis
s t u & R 8 ` n v w X s x o p q r n l m R ( y < o p q r s t l m n R 2 3 g k
,
) *,
5 ( l m z I n { | F o p 2 3 J 0 1 2 3 n ~ } ~ ~ f v w l { J v w > V n ,
` s t v w y & n 4 v w # o p S r < =,
- .Moskowitz
1972
4 å æ × ß n o x u ç ] ~ ï 3 ¤ õ ¹ w è ¹ ~ 3 ¤ , ³ £ $ ã ~ q j ³ u ï Ú accounting-based , À Á ù í ý õ ö Â , Griffin and Mahon 1997 ® Orlitzky et al. 2003 ® Guenster et al. 2005 ® Aigner 2006 ® Nelling and Webb 2008 ³ ª Dam 2006 Â h ` î ì ¹ ~ ® ; t ¹ ,$ ã ~ q j ³ n Ú market-based ,Á õ , Hamilton et al. 1993 ® Guerard 1997a, 1997b ® Brammer et al. 2005a, 2005b ] Anderson and Smith 2006 Â h ` õ $ ã ~ x u l m o p q r s ¡ ! n g k Y ¢ y
67
l m £ 0 t ¤ K n F ¥ w n ~ f ¦ n § ¨,
H + s ¡ ! m g n ¤ K ¨,
L l m ; © n a ª « ¬ m m g; Cochran and Wood
1984
~Moskowitz
1972
n £ 0 t ®,
K | ¯ ) x ° ± D > p | ` ² f ³ 3 m,
H + o p q r n ¡ ! J p ´ 2 3 µ R 5; McGuire et al.
1988
~ 0 ¶ · ¸ Fortune
X s n o p q r £ ¹ t ®,
H + £ ¹ # M 9 k º l m y } n p ´ 2 3,
» J ¼ ½ 2 3 a ¾ « ¬ Q U ¿ À R 5; Waddock and Graves
1997
K | o p q r v w < [Kinder, Lydenberg and Domini
KLD
~ P Á Â } Ã Ä n l m F X s n o p q r £ ¹ t ®,
Å R 5 , Ã J ¯ ) x H +,
£ ¹ Ã Y s M k º l m y } Ã ` n p ´ J ¼ ½ 2 3,
* c z n £ ¹, Tsoutsoura
2004
K | ¯ ) x ° ± Æ Ç H +,
l m < o p q r s t n ¡ ! J L D > « ¬ m F I = « ¬ m J È É « ¬ m µ + M 5 ,,
Ê Ë n v w .Spicer
1978
FChen and Metcalf
1980
FMahapatra
1984
FRusso and Fouts
1997
FThomas
2001
FZiegler et al.
2002
FKing and Lenox
2002
0 l
,
r & v w L s t # Ì 7 4 5 2 3 < =,
- .Vance
1975
,
K |Moskowitz
1972
n £ ¹,
H + o p q r s ¡ ! m g n ,
R m ( L u Í Î t Ï F } Ã Ð R Ê Ë » o p q r s ¡ ! m k n & m k n a ¾ « ¬ m; Newgren et al.
1985
+ , & X s + , £ Ñ t ¨,
H + u c > Ä & X s + , £ Ñ n L ; © a ¾ « ¬ m k; Brammer et
al.
2005a
K | Ò Ó B D v w < [ ethical investment research service
Ô Õy
451
l m F X s n o p q r £ ¹ H +,
£ Ö g × o p 2 3 n + Ö n l m L a ¾ « ¬ m R m k Anginer et al.
2008
K |1983
Ø2006
` { 0 ¶ · ¸ l Ù Ú * ¢ ¦ Û H Ü Ý n l m Þ £ ¹ D ß H +,
Û H Ü Ý n l m L a ª « ¬ m à R k ( L l m á â ã u x o p q r J 6 o p q r n 2 3 i,
ä å + , o p q r y ` n Z c l,
Y \ æ ¨ z I L ` Ã n ç ð ] õ ~ è ; ¹ j ó Ñ Û event study é ê $ ã ë ó ý ì í x u µ Ï î ~ ï Ë õ ,À Á Worrell et al. 1991 ® Clinebell and Clinebell 1994 ® Hannon and Milkovich 1996 ® Posnikoff 1997 ® Wright and Ferris 1997 ® Teoh et al. 1999 ³ ª Brammer et al. 2005b ] Becchetti et al. 2007 Â w F ð ñ Y + f Ê ò l m u Æ S 0 1 ó ô 0 ` Ã p & R 8 N n
,
- . o p q r l m n ; © Æ m N,
Q õ ^ ¹ m ¨ z I n 2 3 i,
Y U V n ö x + } C ( o p q r s t ÷ } n N O ø S + Æ 3 #,
c i,
â ã Ð,
u £ Ñ l m 2 3 n ¯ ) ù Ð = ú û ü Ã o p q r l m t ,
, Q t n > V r p û ü Ã Z ý l = * p ¯ ) Ñ ´ n # > = þ ÿ,
y × t { | þ ÿ x y : Ê ò ä å u o p q r n ` ý & l,
L n ` Ã 1 | } R 0 S w,
× æ F L b R c other things being equal
n ó ô,
5 , - { | þ ÿ 6 Heckman and Robb
1985, 1986
,
{ | þ ÿ s - & } E:
F F " Z æ n à ] { | selection on unobservables
,
× ¯ ) ù Ð n û ü Ã J ÿ G ý & R 5 ;
l,
Q + } C " æ n à ] { | selection
on observables
,
× 7 N O l m o p q r n Ã,
y x t £ Ñ l m 23 ¯ ) ù Ð n H I Ã â ã Ð
, Heckman
1979
n i J V two-stage
me-thod
J = K 4 5 L M Ð n N > V matching method
O P } C Q JÇ F > = n { | þ ÿ i J V n F R + Ñ ´ < m Æ ò
,
æ { | þ ÿ S @ G,
×Inverse Mill’s Ratio,
F T R Q + 9 C y G ] t £ Ñ 2 3 n ¯ ) ù Ð n H I Ã > V & ,
F + @ A g ^ identification
n {,
Y + u l m o p q r n < m Æ ò Ð Ø U & H I Ã J £ Ñ 2 3 ¯ ) ù Ð n ÿ G U 5,
8 : ( V W æ Ã i,
i J > V n p K + ;
F T Q + u v w Ð F £ Ñ n 3 # u I â Ð t o p q r L 2 3 n N O p x t Æ ò = n Z c * & M N n ,
×LaLonde
1986
n X £ Heckman et al.
1997
fHeckman et al.
1998
r tK Ê Ë n " V
,
tHeckman
1979
n i J > V U V Y U " æ n à ] { | F > = n { | þ ÿ 5 ´ ] h ,À Á } ² ~ f ,( j ^ q { Á ´ î ® ´ ñ ] | q î ® ÿ ® ² ® { ® ® ¬ ] © _ î h ~ } ,è ,Ø è , ! " ! ^ " # $ % ~ h Å & ' j ( ) * ¤ + v ¦ ~ f j ( ¡ w 6 ¯ õ 3 ¤ à . / Ö ç ^ " & ' ~ 0 1 î ,É ¦ å ê / « ü Ð Ñ h 2 ~ å á ,« ¡ Û j 3 î 4 x 5 ù í  6 7 ù 8 ³ 9 ò : ; < Å ~ = û > ? ,À Á Cochran and Wood 1984 ,ý j ³ h 2 í o § ~ $ ã ¤ Ö ç 0 1 ,À Á Vance 1975 ] Newgren et al. 1985 w @ N > V s ~ , - á n ±
,
L I Z [ + Ô o p q r n _ l m,
" æ n ` Ã + , R w t \ ,
u y o p q r n l m Ð ] W S ` q ^ _ J N,
p N Ç { u ` Ã n ` o S Y U f o p q r ¨ J 6 o p q r ¨ ¨ l m z I Ð a t N u ` U ,
7 ¨ l m z I { n 2 3 ,
; © } d,
Y s ) x ( o p q r x e > = n N O,
j k å ` Ã n ð ñ 8 Z [ ¦ s ÅRubin
1973
F t K,
s l Q t | | ( = K F 4 5 J u ? $ v L M 8 w ` Ã 8 9 i, Rubin
1973
n > V u x z I N i p > = ^ ` y ! n ,
× { T z I F & w ` Ã © R w Z \,
Y p g ^ F & Ã © R c { n N z I Ã N z Y U Rosenbaum and Rubin
1983, 1985a, 1985b
t K { M Ã N > V propensity score matching method
} H | y } , L I Z [ + u N ^ ¡ Ð Y ` y ! | æ ~ y !
,
F R t ` Ã Ñ ´ < m h Ã,
× o p q r n < m h Ã,
Y + x t i j z I n < m h Ã,
â ã ; t { M à h à propensity score
function
,
Y O z I n ` à k C < m h à Y s æ } z I n Ñ ´ < m,
Y + x t i j z I n Ñ ´ < m,
S { M Ã propensity score
F T R Q + Ô o p q r ¨ Ð n _ z I,
\ 6 { M Ã + , R w t \ ,
u y o p q r ¨ Ð n ] W N z I y > V c i l m { | þ ÿ J z I N i ^ ` y ! n 9 I â n o 9 u £ Ñ < o p q r s t 0 1 2 3 n N O,
c i | | N > V ~ p { | þ ÿ n Õ y q t Ï F London Stock Exchange
7
À Á h ~ ù í : ® \ © } 8 ª h ~ b c  w
8
e ² ; ï f g ~ ù h i ï ý j ë experimental or treatment effect ,¼ ] ï ` î k ï " treatment group ,l ¼ ] ` k Å m " control group ,n o ~ ï å á Å m " ] ï " è é p £ h w ³ 3 À ,z q x u ~ n o î ï " , l z q x u ~ n o ~ 6 7 " ,z q x u Å $ ã ~ > ? r ï w
9
Rubin and Thomas 1992 â õ 8 Å Û = ³ ¡ . Ý ö ÷ ~ w ì Û ~ É Ú h ¯ ,À Á Persson 2001 è c x 5 ] Å £ ~ > ? ; Hutchison 2004 IMF ï Å è c ë ~ > ? ; Glick et al. 2006 ê 2 ù 7 ] (
ä ¡ è ; Vega and Winkelried 2005 ] Crowe 2006 é ê ¢ £ ¤ ¥ ( ¦ e § ~ c ¬ « ï ¾ ( ¦ ] ¨ ² ù 8 ~ © v { ; Ham et al. 2004 ª } ~ } « ¬ Ï ] ù ë ~ è w
J ® ¯ i «
Financial Times
F ° c = ± n < [FTSE,
u2001
` % ± åFTSE
o p q r } Ã FTSE4GOOD Index
,
Y g ^ ² ³ o p q r P Q n l m ' ´ ( } Ã Ð,
~ P B D N µ D - .FTSE All-share
} Ã Ð ¶ { K o p q r · ¸ tFTSE
Õ y o p q r } Ã FTSE4GOOD UK Index
: Y2001
Ø2005
` { ¹ $ t ' ´ (FTSE
Õ y o p q r } Ã Ð n ) Ê t op q r
,
uFTSE All-share
} Ã Ð y t ' ´ (FTSE
Õ y o p q r } Ã n l m Q º t 6 o p q r õ ^ x Ê ò n R 2 3 g k Y s t P < o p q r s t 0 1 2 3 N O 3 # n » B t ¼ ½ Z ¾ J ¿ j o p q r u < o p q r s t Q J < o p q r s t Ç n 2 3 } £ Ñ o p q r 2 3 n N O?
ö x + : U V À _ < l = s t n Á Â i {;
F T F < o p q r Q J o p q r Ç n à A $ v + , Y & O,
* O Z \ ° ±;
F § F â ã Ð > Ä £ Ñ program evaluation
S Å Æ 3 # x policy impact analysis
n x V + K | µ J > Ä J y µ J > Ä n A X s £ Ñ,
- .Heckman et al.
1998
~ f 7 Ç Q u È É Ð t f 5 ( N > V n ` | | v w I â n Ê [ J 7 v w + R c n 5 ( Ë Ì @ Í,
Î Ì t N > V n Ï Ð;
F § Ì Ñ á o p q r à J 2 3 à n ~ ;
F Ò Ì Ó 5 ( o p q r < m h Ã × { M à h à n Ñ ´,
~ f £ Ñ o p q r s t 2 3 N O n > V;
F Ô Ì « Õ I â n s t #,
Ö × { M à h à n Ñ ´ # F z I N n # J ¹ m ~ f z I N Q J N Ç ¨ l m n 2 3 ;
¦ Ç Ì t 2.
Ø Ù Ú Û2.1
Ü Ý Þ ß à á â ßã ä
Dehejia and Wahba
2002
VShen and Lee
2006
,
åY
i1*Y
i0 æ ç è é êi
ë ì ê c d & ' ( ) ë ì V í c d & ' ( ) ë ì î M K ï ð ñ,
ò Y M ð ñ ó ô ê,
τ
i= Y
i1− Y
i0,
õ ö ÷ ø ù ú û ü Q ý þ & ' ( ) ÿ ê 0 ð ñ M ñ î
,
ù M,
$ % c d & ' ( ),
0 H í c d & ' ( ) î M ð ñ ó ô,
τ
i|
Ti=1≡ E(τ
i|T
i= 1) = E(Y
i1|T
i= 1) − E(Y
i0|T
i= 1),
M ñ
average treatment effect on the
treated
,
T
i= 1(0)
éi
ë ì í c d & ' ( ) õ ö ø ! " # $ 2 % & ' c d & ' ( ) M $ % ( ) * + ü , í c d î M ð ñ -,
E
(Y
i0|T
i= 1),
. / ù 0 M ñ 1 D 2 b 1 0 þ 3:
τ
id≡ E(Y
i1|T
i= 1) − E(Y
i0|T
i= 0).
4 4 bτ
d i 5 êτ
i|
T i=1M è 6 7 8 9 M : ; 8 9 < c d & ' ( ) V í c d & ' ( ) M = > Q ? @ 6 7 A B C D,
. ê ( : ; E F,
ò G H I J K L M L N O P Q G ö R,
. /Y
i1, Y
i0⊥T
i,
⊥
é S T,
HE
(Y
i0|T
i= 0) = E(Y
i0|T
i= 1),
J = C U ä V G H W ã V ø ignorability of treatment
A,
X Rτ
d i= τ
i|
Ti=1 Y Z,
< ò G H M = > Q ? @ 6 7 A ó ô [ O î,
ò G H # \ ] ^ H L N O P,
. /E
(Y
i0|T
i= 0) =
E
(Y
i0|T
i= 1), τ
id \ êτ
i|
T i=1M _ ` è 6 7,
þ 3 Mτ
d i I ' Vτ
i|
T i=1a h . ó,
/ ê + , - . ù b 8 9 êRubin
1973
c d M 8 9 S T ) * conditional independence assumption
,
7 e 0 J f ê:
τ
i|
Ti=1≡ E(Y
i1|T
i= 1) − E(Y
i0|T
i= 1)
= E
x{[E(Y
i1|X
i, T
i= 1) − E(Y
i0|X
i, T
i= 1)]|T
i= 1}
= E
x{[E(Y
i1|X
i, T
i= 1) − E(Y
i0|X
i, T
i= 0)]|T
i= 1}
é 1 Q ò G H = D g M ? @ 6 7 A
,
û ü J b h & ' ( ) $ % V V \ & ' ( ) $ % M ð ñ ó ô i ý þ M ÷ ë [ j k 0 ð ñ M ñ2.2
Ü Ý Þ l mRubin
1973
1 2 M n o ( H < > Q ? @ 6 7 p q î,
Q r ò G H (> Q ? @ 6 7 ù D # C s t
,
J D u v G H w + x y z G H 7 O { | } M ~ FRosenbaum and Rubin
1983, 1985a, 1985b
c d æ 7 / 0 2,
I q ; e 0 ' N ; e G H M > Q ? @ 6 7,
J b R ' æ 7 7 b S æ 7:
P (X
i) = P (T
i= 1|X
i) = E(T
i= 1|X
i),
0
P (X
i)
é ( $ % ? @ 6 7 êX
i,
c d & ' ( ) M ,
ê æ 7 ö Rosenbaum and Rubin
1983, 1985a, 1985b
*Rubin and
Thomas
1992
VRubin
1973
8 9 S T ) * M v T J b 2 ? @ 6 7 M g @ æ 7 M g @ ù,
Y
i1, Y
i0⊥T
i|X
i⇒ Y
i1, Y
i0⊥T
i|P (X
i), ∀i;
τ
i|
Ti=1= E
P (Xi)τ
id|
Ti=1,P (Xi)|T
i= 1
,
1 Q ò G H D g M æ 7,
8 9 S T ) * v T,
τ
id= τ
i|
Ti=1 æ 7P (X
i)
M þ 3 J hProbit
Logit
2.3
Ü Ý Þ ò ë ì G H M ? @ 6 7 2 ? @ 6 7 M æ 7 ` L M J D @ C ,
ö r ,
0 & ' ( ) ë ì,
û ü ? @ 6 7 g æ 7 D M \ & ' ( ) ë ì ê / 0 G H : ' p ( H û ü E ¡ ¢ £ ¤,
¥ N ; / 0 Nearest-Neighbor Matching
¦ x § f êNearest
¨ G Hi
] ^ H & ' ( ) ë ì G H ,
G Hj
] ^ H \ & '( ) ë ì y z G H
,
P
i*P
j æ ç ê ò Y M æ 7 / 1 2 0 ; G H,
M y z G H æ 7 © ª D M y z G H ê / 0 G H,
# « ¬ Q d ®,
L N y z G H I J D ' > ¯ ° + / ' C L M G H b 7 e 0 é ,:
C(P
i) = min
j|P
i− P
j|,
C(P
i)
ê ± 2 / 0 p + , V G Hi
æ 7 © D M y z G H ² ³,
( / / 0 ´ ,
/ ² ³ 1 µ ¶ N ; G H ú · * / ² ³ µ ¶ æ 7 ó ô © ¸ MN
; y z G H,
ê1-to-N Nearest Neighbor Matching
4 4 ¥ ¹ ¤ êCaliper Matching
¦ x § êCaliper
,
< º N y z G H V G H M æ 7 ó ô ¸ Hη
î,
|P
i− P
j| < η,
G H I ° + »C(P
i)
M ² ³ ¼ Shen and Lee (2006),
η
½ ¾ ¬ G H æ 7 þ 3 - ¿ ´ ó1
/4,
öC(P
i)
µ ¶ M G H 7 J D1
; b ù4 4 ¥ À ; ê
Mahalanobis Metric Matching
¦ x §Mahala
¨Ma-halanobis Distance:
d(i, j) = (u − v)
TC
−1(u − v),
u
Vv
æ ç ê G Hi
V y z G Hj
M ? @ 6 7 Á,
öC
G H ? @ 6 7 M 6 ô 7 Â Ã 6 ô 7 Ä Å Æ Ç ] È,
) * É ; ? @ 6 7(X
i, i = 1, 2, · · · , 5)
æ ÊA
VB
ë ì,
X
iA, X
Bj, i, j = 1, 2, · · · , 5,
ò ë ì Ë MMahalanobis Distance
ê d(A, B)1×1= X1A− X1B X2A− X2B X3A− X3B X4A− X4B X5A− X5B T 5×1 Var(X1) Cov(X1, X2) Cov(X1, X3) Cov(X1, X4) Cov(X1, X5)
| Var(X2) Cov(X2, X3) Cov(X2, X4) Cov(X2, X5)
| Var(X3) Cov(X3, X4) Cov(X3, X5)
| Var(X4) Cov(X4, X5) | — — — Var(X5) 5×5 X1A− X1B X2A− X2B X3A− X3B X4A− X4B X5A− X5B 5×1 , Ì Í Î Ï Ð Ñ Ò Ó Ô Õ Ö × Ø Ù Ú Û Ü Ý
i
Þ ß à Ü Ýj
ÏMahalanobis
Dis-tance,
á Ñ â ã Ü Ýi,
ä ß à Ü Ýj
å æ ç èMahalanobis Distance
é ê ë Ï Ü Ý ì Ð Ñ Ü Ý í î ï Ì Í Î ð ñ ò ó ô õ ö ÷,
ø ù ú Ù û ü ý þ ÿý Ï í
ý × ì
Mahalanobis Metric Matching with Calipers
Mahala
Caliper
í Ü Ýi
Þ Ü Ýj
ÏMahalanobis Distance
Caliper
(δ)
,
d(i, j) < δ,
ì Ð Ñ Ü Ý íδ
ô õ ö ÷ Ö ! " Ï1
/4,
# Ì Í Î $ %δ,
ûMahalanobis Metric Matching
Þ ô õ ö ÷ Ï Ö & ' ( ) í2.4
Ü Ý Þ l m á * +< £ ¤ / 0 p , - x
,
. / 0 1 & ' ( ) $ % M ë ì 2,
û ü I R ' / 0 x M \ & ' ( ) $ % Hofler et al.
2004
VShen and Lee
2006
,
û ü J b X h 3 ò ¤ 1 0 ] 4 G H / 0 x,
ò G H ? @ 6 7 W 5 ( ó ô b S ó ô ¥ N ¤ 1 0 4 6 $ ) *H
0: (X
T− X
C) = 0,
X
T VX
C æ ç ê / 0 x & ' ( ) ë ì V \ & ' ( ) ë ì M º ; ? @ 6 7,
ö bt
4 7 8 ( G H / 0 x,
? @ 6 7 M 7 ( ò G H Ë M ó ô W 9 F $ 2 : ; 6 $ ) * é ( / 0 x,
] È,
ò G H ( 6 7 ù $ 9 F ó ô < q M ? @ 6 7 ( / 0 x $ 9 F Z ó ô,
é ± p / 0 p x J b < ñ Z = > ò G H Ë M ? @ 6 7 ô @ 4 4 ¥ ¹ ; 1 0 3 (X
IT− X
IC) − (X
AT− X
AC)
(X
IT− X
IC)
× 100,
X
IT*X
IC æ ç è é ( G H / 0 ?,
& ' ( ) $ % V \ & ' ( ) $ % M º ; ? @ 6 7,
ö ( G H / 0 x æ ç bX
AT VX
AC ] é ù 0 3 2 G H / 0 ? ' / 0 x,
º ; ? @ 6 7 M 7 ( ò G H Ë ó ô | } M @ æ A 7 - B O é / 0 B D = > ? @ 6 7 ( ò G H Ë M ó ô @,
ö C ú ( º ¤ / 0 2,
X R O q 7 ? @ 6 7 ó ô | } M @ æ A D < O,
é / 0 p D ñ Z = > ò G H Ë M ? @ 6 7 ô @,
X R ò G H ? @ 6 7 M g < ,
b / æ E ò G H ð ñ ó ô î M + , - . I < ¸3.
F G H I J I K L M N O P Q Ù R S T U V W X Y Z3.1
Ü [ \ ] ^ _ ^ á ` a b c $ % c d & ' ( ) M ? @ 6 7X,
þ 3 æ 7 7,
ë ì M ÷ & ' ( ) ÿ ê M 7 * ,:
P (T = 1) = F (β
X),
ù 0β
ê d 7,
F (·)
, ¿ ´ ¾ i e f 7,
ù b êProbit
,
ö:
X = [1, ASSET, SALES, NETINC, OPERAINC, TURNOVER],
ASSET
ê ! a *SALES
ê g h i j *NETINC
è é k x i l *OPER
AINC
ê m % l [,
öTURNOVER
ê ! a n o ? ¹ Y ê p . q,
˹ Y ê K ï D r . q
,
© x N ; ê s D r . q 6 7 M ¨ b S 0 $ % c d & ' ( ) k M 1 3 H é1,
b È t ä V À ¤ . q M 0 . u p . q4 4
Dierkes and Coppock
1978
*Trotman and Bradley
1981
SFombrun and
Shanley
1990
v d p O M $ % w x ,
y z { | S } ' & ' O ~ M < q
,
V c d & ' ( ) M r < O,
ö M ÷ ë [ R ' M ® ñ J D < O,
. / $ % p V M ÷ & ' ( ) ÿ ê M k d K ï D r . q4 4
McGuire et al.
1988
b SMoore
2001
ê < $ % M K ï ~ F `,
l B q,
I X $ % P ÿ r Z » ! ( ë [ j k ù,
/ ê J ¬ R ! available funds theory
;
N 1 , Posner and Schmidt
1992
b SAlkhafaji
1989
ê $ % ( K ï ~ F _ ` M î ,
I ' $ K H V & ' O ~ M Ã h d
1
¡ ¢ £ ¤
¥ ¦ § ¨ © ª « ¬ ® ¯ ° ± ² ³ ´ ¦ § µ ¶ ¡ ¢
· ª « ¬ ¸ ¹ ¡ ¢ º FTSE All-share» ¢ ¼ ½ ¾ ¿ À FTSEÁ Â
ª « ¬ » ¢ ® ¦ § Ã Ä Å Æ 1, Ç ¾ ¿ À È Å Æ 0 É Ê Ë Ì : · Í Î Ï Ð Í Î +Ñ ° Ò Í +Ó £ Í Î +Ô Õ Í Î + · Ö × Ø Ù Ú Û Ü Ù −Ö Ý Þ ß à á â + ã ä Ë Ì : · å æ Ø ç Ú è Û À −Ú è é Ä −Ú è ê ë ì í £ · · · +è î Û À −è î ï ð −ñ ò å · Ú è ç ó Ú è Û À −Ú è ï ð ì í £ ô õ ö ÷ Ë Ì : · Í Î ø ù ú Ö Ý Ø Ù /Í Î + ¦ § û ü ý þ » ÿ · Í Î ú å æ Ø ç /Í Î × 100% · ú å æ Ø ç / ó × 100% · Ö × ú å æ Ø ç /Ö × Ø Ù × 100% · å æ Ø ç /Ï º î ¢ : FTSE ! " # $ % & ' ( ) * + , - . Compustat Global Vantage / * 0 1 2 3 +4 1 5 6 7 8 9 : # ; < ,= > ? @ A B C D *
opportunism theory
R S I T U V W S X Y Z [ \ ] ^ N _ ` a b c d e fg h R
i K L V W S X
j j
Pava and Krausz
1996
kPreston and O’Bannon
1997
l mStanwick and
Stanwick
1998
n J o p q r s t K L u v m V W a Z [,
J w x y z x {| } W
,
~ N Z [ G r V Y b c a a R S I K L V W a ,
{ J a e R3.2
Ü Þ ¡ ¢ £ á ` a H b ò ¤ 1 0 ý þ $ % & ' ( ) ÿ ê 0 ð ñ M ñ ¤ 1 ¥ H ¦ 3 Á 4,
§ p 3 & ' ( ) $ % V \ & ' ( ) $ % M K ï ð ñ ó ô,
# b ¨ © 2 bootstrap method
ª T « ¬ f Ë b 4 ó ô M 9 F @ , ð ñ ó ô 9 F O H ,
é ] È,
& ' ( ) $ % M ð ñ < ®,
& ' ( ) ÿ ê 0 ð ñ M M,
¯ ³ & ' ° ± ) *;
Y Z ú ð ñ ó ô 9 F ¸ H ,
é ] È,
& ' ( ) $ % M ð ñ é ² < ó,
¯ ³ ³ o ´ o ) * 4 4 ¥ ¹ ¤ 1 0 b = & ' ( ) $ % A V = \ & ' ( ) $ % A ê G H,
b © ¸ 1 2 þ 3 0:
PERFORMANCE
= α + βASSET + λD
CSR+ ε,
PERFORMANCE
ê K ï ð ñ 6 7; ASSET
ê ! a,
h b y z p ñ;
D
CSR ê & ' ( ) 6 µ 6 7,
¶ H1
é G H Ê H & ' ( ) $ %,
¶ H0
é G H Ê \ & ' ( ) $ % & ' ( ) ÿ ê 0 ð ñ M 1 S O ¸ 9 H þ 3 d 7λ,
úλ
9 F O H ,
é & ' ( ) $ % < M K ï ð ñ,
¯ ³ & ' ° ± ) *;
úλ
9 F ¸ H ,
é & ' ( ) $ % K ï ð ñ é ² < ó,
¯ ³ ³ o ´ o ) *4.
· ¸ ¹ º » ¼4.1
Ü ½ ¾ ¿ ^ á À Á Q þ 3 $ % M ÷ & ' ( ) 0 ð ñ M ½ ¾ ' f æ Â Ã $ % c d & ' ( ) M ë ì,
 à C Ä Å M Æ Ç,
Ç ,Bragdon and Marlin
1972
*Folger
and Nutt
1975
VSpicer
1978
È Q l h ë ì H É Ê y z * j k \ ] MË r V j
,
ê c d & ' ( ) M è 6 7,
2,
Ì æ e Y ,McGuire
et al.
1988
*Herremans et al.
1993
b SPreston and O’Bannon
1997
,
§p Í e Î e h M Ï Ð b K Ñ Ò Ó M $ % & ' Ô Õ Ö x ] × Á $ % & ' ( ) 10 q t Ï F J ® ¯ i « ° c = ± n < [
FTSE,
(2001
` % ± åFTSE
o p q r } Ã + , t ' ´ ( } Ã Ð + Ô G Á Ø l n o p q r P Q:
Ù Ú + , # 8 n Û ?;
J K G 5 , H H ! 5 ,;
H I n 5 Â J Ì 7;
Á - Q n P | u ? ~ P;
c i,
o p q r } Ã Ð Y Í ä å 7 t æ F v H ~ f H 0 [ K = n FTSE
u · ± o p q r } Ã i,
° t * F ¢ y F Õ y F J I Ô N : M,
L } Ã Ö å: FTSE
* o p q r } Ã FTSE4GOOD
Global Index
FFTSE
¢ y o p q r } Ã FTSE4GOOD US Index
FFTSE
o p q r } Ã
FTSE4GOOD Europe Index
FFTSE
Õ y o p q r } Ã FTSE4
GOOD UK Index
JFTSE
I o p q r } Ã FTSE4GOOD Japan Index
,
7} Ã O ' ´ å 7 B D : M g ^ Q á o p q r P Q n 11
7 O : M n o p q r } Ã + Å } B D : M * + L M
starting
universe
Ð,
, { K Q á P Q - ý g ! o p q r n · ± * x,
- .
FTSE
u H } Ã FTSE Developed Europe Index
Ð ' ´ g ^ Q Ô G P Q f L o p Ò Ó n { n l m * · ± x Q á nFTSE
o p q r } Ã,
S + Å Ö N u Õ y q t Ï F ¼ l m nFTSE All-share
} Ã Ð ' ´ g ^ o p Ò Ó P Q n l m * · ± xFTSE
Õ y o p q r } Ã % ± 7 o 10 Ü Ý Þ ë 1983 ß à ,á â $ ã n o á ä ~ å Ì ,ä å j " ! Å 8 æ { n o ë } ® ç ç ] è õ é Å « j í o ß 10g n o ~ x u á ä | õ â , ê n o ~ ë ¶ · ; í ¶ ] ë ~ ¶ · ;" ì ; Ë í ù u î ; ï Æ ;ð ñ ® r s ] ò ó } ô ~ û õ ;Å x p ª Ò Ó ö ~ ÷ õ ;¡ ø ù ° n o ù í ,ú û Ö ,á Ö ~ | õ v 0õ ç ð ã ü ý 10õ ç ð þ ó ,û e | 0 Ö × ~ ¬ ÿ 8 r á ¬ n o ~ | 0 ¨ õ w Anderson and Smith 2006 ® Antunovich et al. 2000 r ð á / 1 ÿ | 0 õ 8 ~ n o ,« q è ~ õ h Å 1 ÿ ,@ â Shefrin and Statman 2003 / ~ j h í ~ ® w11
¹ º ~ | , ¡ Kinder, Lydenberg and Domini KLD r s KLD Domini 400x u q 8 KLD Domini 400 Social Index ,j v S&P 500q 8 è v ~ x u { | ö ~ 400¬ $ ã ~ ( õ ë , Waddock and Graves 1997 ] Tsoutsoura 2004 ¦ q 8 é ê n o x u ] ~ è ;Ø Ù , Dow Jones ] p q ù í ë $ ã Sustainable Asset Management, SAM ! ¢ ~ p q { q 8 Dow Jones Sustainability Group Index, DJSGI ," k # { ° | ª $ ã Æ Ç $ í ù p ~ n o j ( p q { ê ;c § ~ ! ç Ý Þ ª " # Ý Þ " ¼ ½ c Ù ¡ ~ | $ ,Å c § n o Ö ç x u å Ì , À Á ! ç Ý Þ 2000 ~ y t % è l & ì ' n o ¸ ® 2004 ~ y l ( $ % n o å Ì ¸ ³ ª 2006] 2007 ~ y n o $ ± å Ì ¸ ;" # Ý Þ 2005 ý 2007 ~ ú ) ~ y n o x u g å Ì ¸ w
p q r } Ã n ö x u ( t P ( o p Ò Ó & { n B D H 7 B D ¨ ~ n
,
c i Y t ? < o p q r s t - Z B D H \ ] n ª _ q [,
Y + w ` } # 8 $ % f o p q r ò B D sustainable and responsible investment
S +social responsible investment, SRI
. n s / 12 I â º o p q r + 3 7 u
2001
Ø2005
` { ¹ $ tFTSE
Õ y o p q r } Ã F ' ´ n l m,
13y } Ã * + L M } Ã =FTSE All-share
} Ã Ð y ¹ t ' ´ (FTSE
Õ y o p q r } Ã n Q Y L º t 6 o p q r Chih et al.
2008
,
( A h ` a,
¨ z I Ð Z Ö × ® ¯ < [ : n v w 4 5 À w u Õ y n P +,
Ç FFTSE
Õ y o p q r } Ã + uFTSE All-share
} Ã Ð õ ^ o p q r P Q F , { K } n,
*FTSE All-share
}Ã > ? Ö × å Õ y q t Ï F ¼ Ï \ n l m
,
~ 7 l m t v w z I Y s , - ° ± y x e;
F T FFTSE
+ q t Ï F J ® ¯ i « n D l m,
u q t Ï F ¼ n l m | } & ¦ ` n å H;
F § FFTSE
¦ s + X s o p q r £ Ñ n : M + q t Ï F ¼ n l m;
¦ Ç FFTSE All-share
} Ã u D ß @ Ð n 0 1 D ß m A @ ) *,
Y 4 5 B C u Õ y ¼ l m n ± +,
U V X s D y n x ~ E Ó o p q r J 0 1 2 3 5 F n D y 4.2
Ü G H ¿ ^ á À Á H H I J M K ø L Æ Ç h b × Á ë ì ð ñ M v ¿ J æ ê ò M,
¥ N ' 3 ê ¥ N,
Ç , ! a O P I J O P ¶,
¥ ¹ b Q R ê ¥ N,
, I S O P T õ ' 3 v ¿ 1 U V ' $ % p i M W X é ²,
YMoore
2001
v d ( × Á $ % & ' ( ) 0 ð ñ M î,
« ¬ ' 3 v ¿ Q A Q R v ¿ ê ® ö McGuire et al.
1988
,
Q R × Á v ¿ T N ¤ ? Z @ M ý þ,
Y s t } ' ! 5 M È % ý \ * E ] 6 k V Q R ^ P M _ k @ 2,
X h Q R v ¿ , I \ O P B C j 2 y z ` 8 O P M . a,
Ç ,Fama and French
1993
*Jegadeesh and Titman
1993
b SPastor and Stambaugh
12
x u 0 í ù j q í ù ö £ í ù ~ î , ½ £ $ ã ~ x u w { ® Ò Ó p q { ª ] " â e 1 ~ ç ð ä , D × j ( 2 ½ £ y ESG¸ , r Ò Ó environment ® x u social ] f ë governance e å < w
13
4 5 ~ $ ã ,î 6 7 ` Á 8 c 9 : British Airways ® Marks & Spencer® ] ® _ < Vodafone Group  w
2003
,
I ' b c G H / 0 î M ¯ Ò H X h ò M ð ñ v ¿ M d n o,
J ã äGriffin and Mahon
1997
H × Á K ï ð ñ î X h M v ¿ ê ! a O P * I J O P * g h O P V I e ,
£ Y D ê ' 3 v ¿ é1
O f 6 7 M ¨4 4 H H M Æ Ç G H
,
bFTSE All-share
v 7 M v g I ê ¥ N,
Ö . / h i * ( ! " Å Ë H j k L l Q M $ % b SCompustat
! " m n n ! " M ë ì,
Ã426
o p q »FTSE All-share
v 7 M $ % µ r & ' ( ) $ %124
,
1 ? ¨,
v M (2001
2005
s Ë t ± °FTSE
u v & ' ( ) v 7 w x M ë ì,
b S \ & ' ( ) $ %302
,
FTSE
All-share
v 7 í t ° w x HFTSE
u v & ' ( ) v 7 M ë ì û ü y ²2000
2005
s Ë M : Ã ë ì M ? @ 6 7 V K ï ð ñ ! " ! " y ² Å Ë ¦ ? z N s ' {2000
s M 0 ., FTSE
(2001
s1
| g ë } u v & ' ( ) v 7 M v g I,
J z w N ~ ° » & ' ( ) v 7 M $ % ? N s ? q ; s ( & ' ( ) ÿ ê ù é ² _ ` M ø Y,
. / û ü ÿ æ E s M ! " 14 l, 124
n o p q r + u2001
Ø2005
` { & ` t ' ´ ( o p q r } Ã ,
u Ç 8 n ) º i x Ð : Y D E o p q r n _ º,
× u2001
Ø2005
` { 7 8 t ' ´ ( o p q r } Ã ,
~ y w º Î n ¨ z I D ß X s £ Ñ l m ` Ã J 0 1 2 3 D ß } C (Compustat
D ß @,
m t ` 5.
5.1
Ü Ý Þ á a b ; ç 6 7 ê ¥ N,
ä V I p À ¿ ´ ó M G H K ê N -,
I G H . ± / x,
& ' ( ) $ %633
; ë ì Â s firm-year
G H o,
\ & ' ( ) ê1,674
; é2
O f / G H * & ' ( ) $ % V \ & ' ( )14
ù ß ! è ~ Ø Ù è e ; j , © ï ~ « è ¯ õ ,å æ ß è Ë ~ ¡ î ¡ õ 8 ¢ 8 ~ . £ ù 8 w
2
¤ ¥ ¦ § ¨ © ª ñ « Ã Ä ª « ¬ ¬ è ® ª « ¬ ¬ è ¯ ° ¢ ÿ ± ² ³ ´ µ ³ ¶ µ ¯ ° ¢ ÿ ± ² ³ ´ µ ³ ¶ µ ¯ ° ¢ ÿ ± ² ³ ´ µ ³ ¶ µ ¦ § µ ¶ ¡ ¢ · Í Î 24.2890 53.8770 0. 0041 375 .0700 31.0480 67. 2950 0. 2060 373.6200 21.7320 47. 6090 0. 0041 375.0700 · Ö × Ø Ù 7.1836 9.3791 0. 0101 57 .8190 10.1920 9. 9776 0. 3094 50.8090 5.9677 8. 8449 0. 0101 57.8190 · å æ Ø ç 0.3309 0.6934 − 3. 7324 4 .8753 0.4782 0. 9223 − 3. 0924 4.5433 0.2751 0. 5744 − 3. 7324 4.8753 · Ú è ç ó 0.6102 0.9184 − 3. 1467 8 .4596 0.9028 1. 1488 − 3. 1467 5.3373 0.4920 0. 7769 − 1. 0135 8.4596 · Í Î ø ù ú 6.6358 10.8200 0. 0974 131 .3500 6.3461 11. 6790 0. 1448 131.3500 6.7529 10. 4560 0. 0974 111.2100 û ü ý þ » ÿ · Í Î ú 3.9948 5.0418 − 16. 3060 26 .1630 3.5635 5. 2182 − 16. 3060 24.7970 4.1581 4. 9655 − 1 4 .3560 26.1630 · ú 12.2400 24.7860 − 376. 5400 449 .6700 10.2760 21. 5650 − 174. 1700 211.4400 12.9890 25. 8770 − 37 6. 5400 449.6700 · Ö × ú 11.2450 9.7627 − 36. 9650 59 .1140 10.2340 8. 7975 − 19. 1070 59.1140 11.6530 10. 1020 − 3 6 .9650 56.5820 · 2.9904 23.2110 − 45. 2900 558 .0200 1.4521 3. 9427 − 10. 1760 58.8650 3.5850 27. 2000 − 4 5 .2900 558.0200 : 8 9 : # 2 · ¸ ¹ º » ¼ ¸ ½ ¾ » ¿ À > ¿ Á 3 Â Ã % Ä Å Æ Ç * 0 È É % 2000 Ê 20 05 Ë * = > Ì = > Ì & Í Î 633 1,674 Ï Ð Ë Ñ Ò Ó *Z [ a Ô Õ Ö × Ø Ù Ú Ö × a Û Ü Ý Þ ß R 15Å n Ð à H +
,
; © } d,
Ô ` Ã Ð ä å D > ³ 3 m l,
o p q r n D > F È É á â F ã Ç á K J % K = R 9 m 6 o p q r g,
t P å Ê ò n ` Ã p u ä ¡ ! n * R t ,
å - } d,
o p q r J 6 o p q r n ; © D > Ã â O t310.48
æ J217.32
æ,
; © ã Ç á K O t4.782
æ J2.751
æ,
; © D > ³ 3 m O t6.3461
J6.7529
Å n2
n Î à H +,
; © } d,
o p q r n Ò 2 3 } ~ R 9 m 6 o p q r } k,
å - } d,
o p q r J 6 o p q r n ; © D > « ¬ m O t3.5635%
J4.1581%,
; © _ a ç è O t1.4521
J3.585
) *,
Ñ á é ´ n # ê ë 5 ( ¿ À n » B ì1
í K z I N Q,
o p q r J 6 o p q r n Ô l m ` Ã n N ì D > J È É á â } d,
o p q r n N u h à _ { m t ; ©,
6 o p q r n î Ã N Q R þ ï,
x y ; © } d,
o p q r n Æ R m N;
ã Ç á K F % K L } d,
o p q r u à _ m k P n î à R m 6 o p q r U,
u à _ m g P n î à Q J 6 o p q r R 8,
x y ; © } d,
o p q r n Z K ® â Ã + R m ` n;
D > ³ 3 m * ð,
ì Ð Q Z \ ñ K ¨ z I { n ò ¿ ì2
w ó ì1
F ô,
u ( n + l m n 0 1 2 3 Ã ) *,
ì Ð : U V õ ö 9 ÷ ø,
; © * ð,
¨ l m z I n Ò 2 3 ù R m g S m k n3
µ + n + o p q r û ü Ã F Ô ` Ã J Ò 2 3 Ã n R 5 , Ã Å F n , Ã H +:
Ç F o p q r û ü Ã J D > F È É á â F ã Ç á K J % K = µ k ! R 5,
å - } d,
J D > F ã Ç á K n R 5 , Ã O t0
.1396
J0
.1201,
* o p q r û ü Ã J D > ³ 3 m Q w U R 5,
R 5 , Ã &−0.0196,
o p q r n Æ m N F Z K ® â m `;
F T F o p q r û ü Ã J Ò 2 3 Ã n R 5 , Ã t _,
L R 5 ( k,
- . J D > « ¬ m F _ a ç è n R 5 , Ã O t−0.0926
J−0.0451
15 $ ã q { ù 8 ~ ù í ® & ' ú û ® ü ú ] ê o % ý _ % þ ÿ b wѨ ᇴ Ѩ ᇴ Ѩ ᇴ Ѩ ᇴ Ѩ ᇴ ዚஐᗝ ྤய ᒉຽӀৈ ඁޢஐӀ ྤயฉᖼத 0 0 10 20 30 40 50 100 150 200 0 0 5 10 15 20 50 100 0 -4 -2 0 2 4 100 200 300 400 0 -4 -2 0 2 4 50 100 150 200 0 0 10 20 30 40 20 60 80 100 : & 1 = > , 1 = > *
1
! " # $ 4 4 2 / 0 ? M % b ¦ 3 & ],
] È,
I ' ' o & ' ( ) $ % M ð ñ é ² 0 < ó M % o õ ö,
2 H ò G H M ? @ 6 7 5 ( < M ó ô,
I : ó ô Ö . = >,
( D ð ñ ó ô ] ^ H M ÷ & ' ( ) ÿ ê M,
} C ? @ 6 7 ó ô ) * M &
3
+ , - . / ª « ¬ ¸ ¹ ¡ ¢ Í Î Ö × Ø Ù å æ Ø ç Ú è ç ó Í Î ø ù ú Í Î ú ú Ö × ú ª « ¬ ¸ ¹ ¡ ¢ 1.0000 Í Î 0.1396 1.0000 Ö × Ø Ù 0.1999 0.7763 1. 0000 å æ Ø ç 0.1201 0.3627 0. 3971 1. 0000 Ú è ç ó 0.2009 0.6038 0. 6505 0. 7644 1. 0000 Í Î ø ù ú − 0.0196 − 0.1419 − 0. 0682 − 0. 0968 − 0. 1383 1. 0000 Í Î ú − 0.0926 − 0.1720 − 0. 1786 0. 3619 0. 1317 0. 0961 1.0000 ú − 0.0459 − 0.0587 − 0. 0648 0. 2476 0. 1053 0. 0657 0.5068 1. 0000 Ö × ú − 0.0623 − 0.0690 − 0. 2159 0. 2371 0. 2293 − 0. 0882 0.5075 0. 2580 1.0000 − 0.0451 − 0.0189 − 0. 0318 0. 0200 − 0. 0093 − 0. 0223 0.0688 0. 0404 0.0193 1.0000 : Ò 1 0 5 1 2 : # ¸ 3 Ï 8 9 : # 4 Ï 5 6 : # 7 8 9 # : ; , È É % 2000 Ê 2005 Ë , < = 2,307 Ï Ð Ë Ñ Ò Ó *Ѩ ᇴ Ѩ ᇴ Ѩ ᇴ Ѩ ᇴ ۵ڌಡ࿌த ྤயಡ࿌த Տ۵࠳ዶ ዚಡ࿌த 0 -20 -10 0 10 20 30 50 100 150 0 -400 -200 0 200 400 200 400 600 0 -40 -20 0 20 40 200 400 600 0 -40 -20 0 20 40 60 50 100 150 :> ? 1*
2
@ A B C ! " # $5.2
Ü Ý Þ ¤ 1 þ 3 & ' ( ) $ % V \ & ' ( ) $ % M æ 7,
ö ± p G H / 0 x I J R ' / 0 x M \ & ' ( ) $ % G H é4
O f / bProbit
þ 3 æ 7 7 M & ¥ N D 9 ( N,
! a M þ 3 d 7 9 F ê E(−0.0049),
g h i j M þ 3 d 7 9 F ê (0.0227),
ö m % l [ M þ 3 d 7 9 F ê (0.2363),
: é & ' ( ) ë ì M ! a p < ¸ * g h j < q * l j < q õ ö,
2 1 ? M d 7 Ä Å F ²,
× Á p . q M ! a V g h i j Ë 5 ( d 7 ê0
.7763
,
L G M E G F h ( × Á l ~ F M k x i l V m % l [ d 7 ê0
.7644
ù,
: ¤ E G s t a h H I 6 7 Ë Ã J @ M K J ö X þ 3 & C L ä V ( p . q V K ï . q ù M ^ X h N ; 6 7,
ö N I è é s D r M ! a n o µ r ( ¼,
. ö X * M ³ b c / £ ¤,
æ ç 9 ( é4
M ¹ É
4
O P Q R S © TProbit
U V W X Y Z [ \ ] [ \ ^ [ \ _ [ \ ` [ \ a −0.8059 −0.6947 −0.7530 −0.7790 −0.8030 b Z c (−15.7) (−14.6) (−15.2) (−15.4) (−15.6) −0.0049* 0.0071*** 0.0022 d e (−1.76) (3.88) (1.02) 0.0227*** 0.0244*** 0.0173*** f g h i (3.84) (6.42) (3.78) −0.0969 0.1441*** 0.0859* j k l m (−1.27) (2.86) (1.69) 0.2363*** 0.2408*** 0.1559*** n o p q (3.39) (5.39) (3.35) 0.0004 0.0008 0.0015 0.0000 0.0010 d e r s t (0.11) (0.24) (0.47) (0.01) (0.30) u v w x y Z z 1,501 1,501 1,501 1,501 1,501 PseudoR-square 0.0415 0.0200 0.0318 0.0343 0.0390 { | :} [ \ ~ W X pooled ~ Z W X Y Z ty , * ** *** W X Y Z 10% 5% 1%~ b Z c ¡ ¢ £ ¤ ¥ ¦ § ¨ © ª «,
¬ ® ¯ ° ± ² ¥ ¦ § ³ ´ µ ¶ · ¸ ¹,
º » ¸0
.0071
®0
.1441,
¼ ¬ ½ ¾ ¿ ¥ ¦ § ³ ´ ¸ ¹,
À Á ¶ · Â ¢ £ Ã Ä Å Æ Ç ² È ¥ ¦ § ³ ´ ¶ · ¸ ¹(0.2408)
Â É ¢ £ Ê ® ¢ £ Ë Ì,
Í Î Ï ¢ ® Ð ² Ñ Ò ¥ Ó ´ Ô ¦ § ³ ´ µ ¶ · ¸ ¹,
¬ ½ ¾ ¿ ¥ ¦ § ³ ´ Õ Ö × Á ¶ · Â ¡ Ø Ù Ú Û Ï ¢ Ü Ý Þ Ð ² Ñ Ò ß à á Ù â ã ä å æ ç è ä é ê ë ì í î ï ð ¸ ¥ ñ ò Ü ó,
Ü ô õ ö É ¢ £ ¤ Þ ¢ £ Ê ® ¢ £ Ë ÷ Ì ø ù ú Â û ü ¥ ý þ ÿ ¬ Í Î Ç ¥ Ï ¢,
Ü ô ß ¢ £ ¤ ¥ ¦ § ¨ © § ç è ¥ º ´ ß à ð Â ë Û ÿ Ô ¢ £ ¥ ¦ § ¨ © ð í ¥ º ¨ © Â ¦ § º ´ ´ °,
ç è ¥ º ´,
· ² Ê ð Â5
! " É ® °,
Ë # $ % Ó ´ É & ' ( ¥ ) * ´ + , - . ¨ © Â õ ö ª «,
& ' $ % Ó ´ ¥ + , % É ® ° Å / ¶ 0 1 Ó Â 2 3 Þ É,
& ¥ Ê # $ % Ó ´ ¬ Þ 4 5 ± 6 Þ ¯ ° ± ² ® Æ Ç ² È 7 Å
5
8 9 : ; < = > 8 9 : ; < = ? @ A B T C D E C D F G ¥ H I Ã Ä G ¥ J Ne arest-N e ighb or Matchin g (N e arest) C a li p e r Matchi ng (C a lip e r) Mahalano bis Me tric Matchin g (Mahal a ) Mahalano bis Metri c Matchi ng wit h Cali pers (Mahala Calip e r) K L M K L N O K L M K L N O K L M K L N O K L M K L N O K L M K L N O P Q R P Q R (t S ) P Q R P Q R (t S ) P Q R P Q R (t S ) P Q R P Q R (t S ) P Q R P Q R (t S ) 5. 933*** − 0. 517 − 0. 601 0.217 0.099 T U 14.83 8 .895 1 4 .8 3 15. 35 14.48 15. 08 14. 83 14. 61 7. 451 7. 35 3 (5 .50) (− 0.37) (− 0.44) (0 .16) (0 .15) 4. 224*** − 0. 222 − 0. 328 0.636 − 0. 083 V W X Y 10.19 5 .968 1 0 .1 9 10. 41 10.01 10. 34 10. 19 9. 556 5. 484 5. 56 7 (8 .07) (− 0.30) (− 0.44) (0 .91) (− 0.17) 0. 189*** 0.022 0.027 − 0. 002 0.003 Z [ \ ] 0.445 0 .256 0 .445 0. 423 0.448 0. 422 0. 445 0. 447 0. 263 0. 26 0 (4 .82) (0 .40) (0 .50) (− 0.04) (0 .10) 0. 411*** 0.102 0.101 0.092 − 0. 017 ^ _ ` a 0.903 0 .492 0 .903 0. 801 0.892 0. 791 0. 903 0. 811 0. 473 0. 49 0 (8 .01) (1 .42) (1 .43) (1 .27) (− 0.38) − 0. 407 0.387 0.402 0.319 − 0. 030 T U b c d 6.346 6 .753 6 .346 5. 959 6.376 5. 974 6. 346 6. 027 5. 145 5. 17 4 (− 0.66) (0 .54) (0 .55) (0 .41) (− 0.06) : Ò 1 e 5 Ñ Ò f g h f g ¾ , = > Ì = > Ì 3 8 9 : # i j # k l m 7 n o * 0 N eares t f g p q r s Ï = > 3 t u & # v 7 w = > Ñ Ò $ % f g Ñ Ò ; C aliper x y z = > = > t u & # 3 k l { | } Ï ~ -$ % Ñ Ò f g ) ; M ahala x y % q r = > 3 M ahalanobis v { $ % f g Ñ Ò ; Maha la Ca lipe rs z M ahala n obis { | } Ï ~ -$ % Ñ Ò f g ) * f g h , = > Ì = 633 Ï Ñ Ò Ó , = > Ì = 1,674 Ñ Ò Ó , p h 4 f g y , ? @ = > Ì & Í % 423 ¸ 420 ¸ 423 z 253 Ï Ñ Ò Ó * 4 2 % Ì Ñ Ò 8 9 : # k t , * ¸ ** *** & Í 1 5 k l p 10% ¸ 5% 1% ) e *Ѩ ᇴ Ѩ ᇴ Ѩ ᇴ Ѩ ᇴ Ѩ ᇴ ዚஐᗝ ྤய ᒉຽӀৈ ඁޢஐӀ ྤயฉᖼத 0 0 10 20 30 40 10 20 30 0 5 10 15 20 25 0 5 10 15 20 0 -4 -2 0 2 4 20 40 60 80 100 0 -4 -2 0 2 4 20 40 60 80 0 0 10 20 30 40 40 20 60 :> ? 1*
3
Nearest
" # 9 F M ó ô @,
(Nearest
/ 0 ,
ö ,
: £ ; ? @ 6 7 M 9 F ó ô @ W /;
¥ ¹ * M,
(Caliper
*Mahala
VMahala Caliper
/ 0 d ² M g M &,
ö ,
ò G H M É ; ? @ 6 7 ª D ¶ 1 ? c S ú Q I ò M $ % M ð ñ > . H M ÷ & ' ( ) : N . q,
© ` X R ò G H M ? @ 6 7 Á D,
. / ( £ ¤ / 0 ,
{ ' / : ; Q r,
I ñ = > / + , - . M K J
6
A B Q G ¥ H I Nearest-Neighbor Matching (Nearest) Caliper Matching (Caliper) Mahalanobis Metric Matching (Mahala) Mahalanobis Metric Matching with Calipers (Mahala Caliper) T U 91.3 89.9 96.3 98.3 V W X Y 94.7 92.2 84.9 98.0 Z [ \ ] 88.2 85.8 98.7 98.5 ^ _ ` a 75.2 75.4 77.7 95.9 T U b c d 5.00 1.20 21.7 92.7 70.9 68.9 75.9 96.7 : Ò 1 3 Ï 8 9 : # p ¡ Ñ Ò f g ¾ , i j # p = > = > 3 É k l 9 ¢ £ ¤ & ¥ * 0 8 7 f g y ¦ ¤ & ¥ ; D ,1 5 ¡ § x y ¨ © Ñ Ò f g ¾ , ' ª = > 8 9 : # k l 9 ; « ,f g ; ¬ * j j ®3
¯ ° ±Nearest
² Y ³,
´ µ ¶ · ¸ ¹ p Ô Õ Ö × a º ² ® R Ø ®1
r » ¼ ½ ¾ ¸ ¿ ± À ´ µ ¶ · ¸ Ô Õ Ö × a º ² Ö Á Â J Ã Ä Å ½ Æ Ç È,
É Ê Ë Ì ¸ Í a Î ¾ Ï Ð Ñ w R 16 n6
« Õ å Å z I N Q Ø z I N Ç,
o p q r J 6 o p q r ` Ã n ; © Ã Y U Ò ¹ Ò ¹ Ö N n Ó z I N Ö - j k ¨ z I u ` Ã n & Ô n # +,
F FMahala Caliper
+ ¦ & 3 n N > V,
x t L N z 9 j k ¨ z I { ` Ã n ,
Ô Ã Y U Ò ¹ n ; © Ã t96.7%;
F T FMahala
FNearest
JCaliper
N n & 3 Î j,
Ô ` Ã Y U Ò ¹ n ; © Ã O t75.9%
F70.9%
J68.9%;
F § F m & n N > V 6 Õ & x I,
å - } d,
u N ¦& 3 n
Mahala Caliper
N Î,
N Ç n 6 o p q r ¨ n z I p253
,
Ö × R 8 ` n C Å !,
) * u N & 3 ¡ ! m k n > V ×Nearest
JCaliper
N Î,
6 o p q r ¨ - & m ` z I 5,
O t423
J420
x y @ A } d,
¹ m & 3 n N > V & N n o C Å ! n Ñ ´ 3 m ,
s Ø ¡ ! m k n N > V Ù Z p n o Ú ` C Å ! Å ( â ã Ð ê ë 5 ( 16 Û Ü Ý Þ , ß à « ü á Å â ã ç / ~ ^ n o " « q { ù 8 ] q ~ ä å ³ Ü æ wç è 5 , ¦ é O n Ó
,
: Y « Õ Ò N V Î n 2 3 ¹ m J ¯ ) x # 5.3
Ü G H * + é7
O f / & ' ( ) $ % V \ & ' ( ) $ % £ ; ð ñ v ¿ ; ç 7 ó ô M æ E &,
M / 0 ? V / 0 x M G H ( G H / 0 ?,
¥ ¹ D,
] È,
& ' ( ) $ % V \ & ' ( ) $ % M £ ; ð ñ v ¿ ó ô D 9 F ê E,
Ç , & ' ( ) $ % M ! a O P V I J O P 0 < \ & ' ( ) $ % >0.936%
V2.789%,
é ] È & ' ( ) $ % M ð ñ é ² 0 < ^ M 4 4 % & é7
¥ À D ¥ ê D M ð ñ A < & F ²,
E F / ë 6 ¤ 1 * (Nearest
/ 0,
< ð ñ 6 7 ê I e î,
& ' ( ) $ % V \ & ' ( ) $ % M ó ô ê E,
é & ' ( ) $ % M I e 0 < >,
+ À ; ð ñ v ¿ T õ é ² d & ' ( ) $ % 0 < > M E G,
Y # C 9 F;
¥ ¹ * (Caliper
/ 0,
& ' ( ) $ % M I e é ² 9 F < >,
+ M ð ñ v ¿ V \ & ' ( ) $ % ì 9 F Z > ç,
VNearest
M g M &;
¥ À * (Mahala
VMahala Caliper
/ 0,
T õ d ² / & ' ( ) $ % M I J O P * g h O P V I e 0 < > M E F,
Y ¦ 3 ù # C í 9 F @ 4 4 î4
ï d (Nearest
/ 0,
ò G H M £ ; ð ñ 6 7 æ / î V î2
M g,
2 H ò G H ð ñ 6 7 M æ / D,
. / $ 2 b ð G ñ ò Z 7 8 ö ,
ò ë ì G H M £ ; ð ñ v ¿ ó Y 0 < < > 4 4 2 b ù æ E & R w,
G H / 0 ?,
& ' ( ) $ % ( £ ; ð ñ v ¿ M é ² ù D 9 F < ó,
G H / 0 x,
9 F < ó M L | } /,
E d ² (Nearest
VCaliper
/ 0 î,
& ' ( ) $ % 9 F < > M I e õ ö,
X § p G H / 0,
N õ C ' & ' ( ) $ % ð ñ é ² < ® M L ,
. /,
ø L & ' o ³ o ´ o ) * 4 4 ª F û ü l h ô æ E ý þ & ' ( ) $ % V \ & ' ( ) $ % M 0 ð ñ >,
È Q ý þ ? b ô 0 M & ' ( ) 6 µ 6 7 þ 3 d 7 M E S 9 F @ é8
O f / / 0 ? V / 0 x M G H ÿ M ô æ E &
7
8 9 : ; < = > 8 9 : ; < = ? @ A B T C D E C D F Ã Ä G ¥ J G ¥ H I Nearest-Neighbor Matching (Nearest) Caliper Matching (Caliper) Mahalanobis Metric Matching (Mahala) Mahalanobis Metric Matching with Calipers (Mahala Caliper) −0.936*** −0.436 −0.419 0.074 0.213 T U õ ö d (−3.13) (−1.02) (−0.98) (0.44) (1.09) −2.789* −0.803 −0.933 −2.877 −0.624 ÷ ø õ ö d (−1.80) (−0.29) (−0.39) (−0.82) (−0.38) −1.419** −0.717 −0.697 −0.470 −0.428 V W õ ö d (−2.55) (−0.81) (−0.84) (−0.79) (−0.56) −2.656* −4.540*** −4.576** −0.249 −0.253 ù ÷ \ ] ú (−1.80) (−2.60) (−2.39) (−0.42) (−0.31) : Ò 1 p Ñ Ò f g h f g ¾ , = > = > i j 5 6 k l * 0 4 2 % i j # k l 3 Z~ ,¿ û ü ý y 1,000þ ÿ , *¸ ** ***& Í 1 5 i j # k l p 10%¸ 5% 1% ) e * Ѩ ᇴ Ѩ ᇴ Ѩ ᇴ Ѩ ᇴ ۵ڌಡ࿌த ྤயಡ࿌த Տ۵࠳ዶ ዚಡ࿌த 0 20 40 60 -20 -10 0 10 20 30 0 50 100 200 150 -200 -100 0 100 200 0 20 40 60 80 100 -5 0 5 10 15 -10 0 -40 -20 0 20 40 60 20 40 60 :> ? 1*4
@ A B CNearest
" #/ 0 ? M & 9 (
Panel A,
û ü F ² < ð ñ 6 7 ê I J O P * g h O P V I e î,
& ' ( ) 6 µ 6 7 M þ 3 d 7 9 F ê E,
æ ç ê−2.5536
*−1.2349
V−2.0687,
ö < ð ñ 6 7 ê ! a O P î,
T õ 6 µ 6 7 M þ 3 d 7 ê E,
Y # C 9 F : é & ' ( ) $ % ( I J O P * g h O P V I e M é ² ù 0 < ó M 4 4 < û ü l h / 0 x M G H 5 æ E î,
E F / Ã M ë 6Panel B
Panel E
æ ç O f / £ ¤ / 0 2 æ ç êNearest
*Caliper
*Mahala
VMahala
Caliper
M ô þ 3 & ¤ 1 * (Nearest
/ 0,
< ð ñ v ¿ ê g h OP V I e î
,
& ' ( ) 6 µ 6 7 M þ 3 d 7 9 F ê E,
æ ç ê−1.3957
V−4.3862,
é & ' ( ) $ % 0 < > M g h O P V I e ;
¥ ¹ * (Caliper
/ 0,
< ð ñ 6 7 ê g h O P V I e î,
6 µ 6 7 M þ 3 d 7 9 F ê E,
< ð ñ 6 7 ê ! a O P V I J O P î,
6 µ 6 7 M þ 3 d 7 ê E Y C 9 F,
VNearest
/ 0 L M &;
¥ À * (Mahala
/ 0,
C ð ñ 6 7 ê ó,
6 µ 6 7 þ 3 d 7 D C 9 F,
é & ' ( ) $ % M ð ñ C ® # í { ' ¦ 3 9 F;
¥ £ * (Mahala Caliper
/ 0,
& ' ( ) $ % M £ ; ð ñ v ¿ D > H \ & ' ( ) $ %,
õ ö 1 < ð ñ v ¿ ê I J O P î,
: G M ó ô ( í ¦ 3 ù M 9 F @4 4 2 b ù æ E & R w
,
G H / 0 ?,
& ' ( ) $ % ( À ; ð ñ v ¿ M é ² ù í 9 F M 0 ^ _,
G H / 0 x,
é ² C ® M 9 F L | } /,
(Nearest
VCaliper
/ 0 î,
& ' ( ) $ % ( g h O P V I e ù é ² 9 F < ó,
(Mahala Caliper
/ 0,
& ' ( ) $ % M I J O P 9 F < > õ ö,
( £ ¤ / 0 2 / £ ; ð ñ v ¿ M16
; 6 µ 6 7 þ 3 d 715
; ê E,
# í d ² 9 F ê E G,
: é & ' ( ) $ % ( Ì M ð ñ v ¿ é ² ù } C ' A \ & ' ( ) $ % ] R ` . /,
X § p G H / 0,
N õ d ² & ' ( ) $ % é ² 9 F < ó M Ì æ L ,
L î C ' ð ñ é ² < ® M E G,
ô æ E M & ' o ³ o ´ o ) * 4 4 é9
O f M ¼ s V é8
g,
ó ô ( H ô 0 b * / î Ë M 6 µ 6 7 b y z î Ë _ % & ( G H / 0 ? M æ E &Panel A
J w,
ð ñ 6 7 ê I J O P * g h O P V I e î,
6 µ 6 7 M þ 3 d 7 9 F ê E,
æ ç ê−2.5632
*−1.2661
V−2.0631,
& ' ( ) $ % ( : À ; ð ñ 6 7 9 F Z 0 < ó q ^Panel B
ØPanel E
F « Õ n # H +,
Ç F uNearest
N
8
8 9 : ; < = ?
? Í Î ú ú Ö × ú Panel A.Ã Ä G ¥ J 4.6332 13.390 11.930 3.7804 · ¢ (32.8) (18.7) (37.3) (5.50) −0.0218 −0.0179 −0.0311 −0.0094 · Í Î (−10.3) (−1.68) (−2.34) (−0.87) −0.3914 −2.5536** −1.2349** −2.0687* · ª « ¬ ¸ ¹ ¡ ¢ (−1.53) (−1.97) (−2.20) (−1.68) Panel B. Nearest-Neighbor Matching (Nearest)4.5310 10.622 12.076 6.1511 · ¢ (17.1) (9.12) (22.6) (3.83) −0.0166 −0.0074 −0.0301 −0.0113 · Í Î (−4.87) (−0.50) (−1.82) (−0.49) −0.4525 −0.1139 −1.3957** −4.3862** · ª « ¬ ¸ ¹ ¡ ¢ (−1.27) (−0.07) (−2.11) (−2.04) Panel C. Caliper Matching (Caliper)
4.5860 10.883 12.119 6.1940 · ¢ (17.3) (9.37) (22.7) (3.84) −0.0165 −0.0071 −0.0323 −0.0111 · Í Î (−4.86) (−0.48) (−1.93) (−0.48) −0.5102 −0.3859 −1.4062** −4.4360** · ª « ¬ ¸ ¹ ¡ ¢ (−1.44) (−0.25) (−2.12) (−2.05) Panel D. Mahalanobis Metric Matching (Mahala)
3.9968 13.888 10.568 1.6899 · ¢ (16.3) (7.01) (21.4) (7.95) −0.0149 −0.0093 −0.0058 −0.0002 · Í Î (−4.73) (−0.37) (−0.37) (−0.05) 0.0312 −3.3210 −0.2484 −0.2335 · ª « ¬ ¸ ¹ ¡ ¢ (0.09) (−1.25) (−0.40) (−0.82)
Panel E. Mahalanobis Metric Matching with Calipers (Mahala Caliper)
4.2088 12.886 10.515 1.6176 · ¢ (14.2) (10.7) (18.6) (5.97) −0.0143 −0.0035 0.0093 −0.0003 · Í Î (−4.63) (−0.28) (0.47) (−0.09) −0.2001 −2.5003* −0.4194 −0.1576 · ª « ¬ ¸ ¹ ¡ ¢ (−0.54) (−1.67) (−0.60) (−0.47) â : ç Ù ³ ~ 4 , ¬ Û ï ë x u n o Å ~ > ? ,ù Ë v 2000ý 2005 w 3 î ñ 4 : PERFORMANCE= α + βASSET + λDCSR+ ε, « PERFORMANCE ù 8 ; ASSET ù í , ³ 6 7 : ;DCSR x u ù 8 , 1ç ¦ " k x u n o , 0ç " k x u n o w " Å ß ,x u n o ú 633e " , x u n o ú 1,674e " ;
Nearest® Caliper® Mahala] Mahala Caliper Å ç , x u n o õ 423® 420® 423] 253e " w
v ï è 8 tî ,â *® **] *** ù 8 ~ ï è 8 Ú õ ç ï è 8 10%® 5%] 1%~ ¡ ç ) * ' w
9
8 9 : ; < = ?
? T ! " # $ Í Î ú ú Ö × ú Panel A.Ã Ä G ¥ J 4.6240 12.995 16.578 4.8095 · ¢ (2.85) (1.58) (4.49) (0.58) −0.0216 −0.0173 −0.0307 −0.0098 · Í Î (−10.3) (−1.62) (−2.31) (−0.90) −0.3949 −2.5632** −1.2661** −2.0631* · ª « ¬ ¸ ¹ ¡ ¢ (−1.56) (−1.99) (−2.26) (−1.67) Panel B. Nearest-Neighbor Matching (Nearest)4.3262 10.224 15.206 3.9296 · ¢ (1.81) (0.97) (3.49) (0.27) −0.0164 −0.0069 −0.0288 −0.0124 · Í Î (−4.86) (−0.47) (−1.73) (−0.54) −0.3073 0.3763 −1.2610* −3.9071* · ª « ¬ ¸ ¹ ¡ ¢ (−0.87) (0.24) (−1.90) (−1.81) Panel C. Caliper Matching (Caliper)
4.3711 10.441 15.244 3.9690 · ¢ (1.84) (1.00) (3.50) (0.27) −0.0163 −0.0067 −0.0309 −0.0119 · Í Î (−4.86) (−0.45) (−1.85) (−0.52) −0.3644 0.1018 −1.2755* −3.9633* · ª « ¬ ¸ ¹ ¡ ¢ (−1.03) (0.07) (−1.91) (−1.83) Panel D. Mahalanobis Metric Matching (Mahala)
3.5095 10.243 6.1073 0.5817 · ¢ (1.42) (0.51) (1.36) (0.27) −0.0148 −0.0088 −0.0034 0.0000 · Í Î (−4.75) (−0.35) (−0.22) (0.00) −0.0158 −3.6456 −0.2758 −0.2640 · ª « ¬ ¸ ¹ ¡ ¢ (−0.05) (−1.37) (−0.45) (−0.93) Panel E. Mahalanobis Metric Matching with Calipers (Mahala Caliper)
3.7141 9.1479 6.0236 0.5222 · ¢ (1.57) (0.96) (1.38) (0.24) −0.0142 −0.0032 0.0106 −0.0001 · Í Î (−4.66) (−0.26) (0.53) (−0.04) −0.2282 −2.6235* −0.3762 −0.2029 · ª « ¬ ¸ ¹ ¡ ¢ (−0.62) (−1.76) (−0.54) (−0.60) â : ç Ù ¬ Û ï ë x u n o Å ~ > ? w ù Ë v 2000ý 2005 w 3 î ñ 4 : PERFORMANCE= α + γTD + βASSET + λDCSR+ ε,« TD î ù 8 , ³ 6 7 î % & Û Ü Ý Þ ,î ù 8 ~ è 8 ï ® ¥ l ' ð w
Î
,
2 3 } ~ t È É « ¬ m J _ a ç è i,
û ü Ã n Ñ ´ , Ã ¿ À t _,
O t−1.2610
J−3.9071,
* uCaliper
N Î r K + Ê Ë n #;
F T F uMahala
N Î,
o p q r û ü Ã n Ñ ´ , Ã u Ò 2 3 } ~ Î Z ¿ À;
F § F uMahala Caliper
N Î,
o p q r n Ò 2 3 } ~ k ( 6 o p q r ,
) * & 8 2 3 } ~ t a b « ¬ m i ( ý & ¿ À F Ò F16
û ü Ã Ñ ´ , Ã N t _,
n Ó N n Ñ ´ # { M Ì 7 o p q r n 2 3 + R m k n Å ~ x # s À,
u ¯ ) ù Ð D E 8 = i { û ü à Z p . s t # { M Ì 7 4 5 2 3 < = n ) * u { M à h à n Ñ ´ i,
: p | n Æ ò = L H I Ã Ö × D > F ã Ç ç è J D > ³ 3 m,
- D E p | c l Ã,
Y + K | 8 l n ` Ã + Ñ ´ l m 8 l t o p q r n < m ) * + , & s - l m tFTSE
' ´ ( o p q r } Ã n < m J ^ + n 2 3 & 5?
17 - . l m ^ + ` n Z K ó ô ( ,
t t ¹ m / , o p q r * * C o p q r } Ã Ð . # + z,
Q : | D E Y l m ^ + n 2 3 x t { M à h à Рn H I à : D E Y Ú l m Q ` ! n ã Ç á K Þ x t Ñ ´ { M à h à i n H I Ã,
* L n H I à - t 8 l n D > J D > ³ 3 m h à n Ñ ´ #,
: w X s z I N,
* n10
« Õ å N Q n z I ~ f Ò z I N Î n ¯ ) x Ñ ´ # u y - ) D E 8 = i { û ü Ã,
x y z I N Q n s t # Y J n9
z I N Q n s t # A * R c q ^ Ò N Î n x #,
× q ^Panel B
ØPanel E,
H + J n10
K + 7 . F F uNearest
N Î,
. ) u F & n 2 3 } ~ Î,
o p q r û ü Ã n Ñ ´ , Ã t _,
» & u D > « ¬ m Î + ¿ À n,
, Ã t−0.6477,
* Ê Ë n / W K + uCaliper
N Î n x # Ð;
F T F uMahala
N Î,
Z 2 3 } ~ t ù,
o p q r û ü Ã Z ¿ À;
F § F uMahala Caliper
N Î,
o p q r n Ò 2 3 } ~ k ( 6 o p q r ,
) * & 8 2 3 } ~ t a b « ¬ m i ( ý & é ´ ¿ À F Ò F u16
û ü Ã n Ñ ´ , Ã Ð &15
K + _ Ã,
c i Y y H = ¿ À t n / W,
n Ó o p q r n 2 3 n + Ø U k ( 6 o p q r x y,
. ) s t # K + 7 8 . ,
17 ñ ¯ õ ,å æ | } | } å æ ~ q ] w
10
8 9 : ; < = ?
? T ! " # $ 0 O P Q R 1 2 3 E 4 5 6 7 Í Î ú ú Ö × ú Panel A.Ã Ä G ¥ J 4.6240 12.995 16.578 4.8095 · ¢ (2.85) (1.58) (4.49) (0.58) −0.0216 −0.0173 −0.0307 −0.0098 · Í Î (−10.3) (−1.62) (−2.31) (−0.90) −0.3949 −2.5632** −1.2661** −2.0631* · ª « ¬ ¸ ¹ ¡ ¢ (−1.56) (−1.99) (−2.26) (−1.67) Panel B. Nearest-Neighbor Matching (Nearest)4.1525 8.9772 7.0924 2.2044 · ¢ (1.62) (0.87) (1.53) (0.22) −0.0157 −0.0082 −0.0173 −0.0043 · Í Î (−4.81) (−0.63) (−1.01) (−0.30) −0.6477* −2.3868 −1.0789 −1.8302 · ª « ¬ ¸ ¹ ¡ ¢ (−1.71) (−1.57) (−1.54) (−1.23) Panel C. Caliper Matching (Caliper)
4.1902 8.7290 7.1161 2.2191 · ¢ (1.64) (0.86) (1.54) (0.22) −0.0157 −0.0079 −0.0175 −0.0043 · Í Î (−4.84) (−0.61) (−1.03) (−0.30) −0.6848* −2.1429 −1.0993 −1.8446 · ª « ¬ ¸ ¹ ¡ ¢ (−1.81) (−1.43) (−1.56) (−1.24) Panel D. Mahalanobis Metric Matching (Mahala)
3.4623 10.307 6.0695 2.0036 · ¢ (1.39) (0.62) (1.32) (0.20) −0.0151 −0.0080 −0.0009 −0.0037 · Í Î (−4.77) (−0.38) (−0.05) (−0.26) 0.0353 −3.7200 −0.2709 −1.6367 · ª « ¬ ¸ ¹ ¡ ¢ (0.10) (−1.53) (−0.39) (−1.11)
Panel E. Mahalanobis Metric Matching with Calipers (Mahala Caliper)
3.9714 15.469 6.3116 3.4131 · ¢ (1.60) (0.88) (1.37) (0.31) −0.0143 −0.0005 0.0056 −0.0018 · Í Î (−4.46) (−0.02) (0.26) (−0.11) −0.4843 −8.9801*** −0.5988 −3.0711 · ª « ¬ ¸ ¹ ¡ ¢ (−1.12) (−2.94) (−0.72) (−1.60) â : ç Ù ¬ Û ï ë x u n o Å ~ > ? w " Å ! ñ , õ 8 ¢ 8 ~ . £ ù 8 ß è Ë ~ ¡ w ù Ë v 2000ý 2005 w 3 î ñ 4 : PERFORMANCE= α + γTD + βASSET + λDCSR+ εw " Å ß , x u n o ú 633 e " , x u n o ú 1,674 e " ; Nearest® Caliper® Mahala] Mahala Caliper Å ç , x u n o õ 337® 336® 337] 195e " w