'LVFRYHULQJ&OXVWHU%DVHG/RFDO2XWOLHUV

=HQJ\RX+H  ;LDRIHL;X6KHQJFKXQ'HQJ
'HSDUWPHQWRI&RPSXWHU6FLHQFHDQG(QJLQHHULQJ+DUELQ,QVWLWXWHRI7HFKQRORJ\+DUELQ35&KLQD
5HFHLYHG;;UHFHLYHGLQUHYLVHGIRUP;;

$EVWUDFW
,QWKLVSDSHUZHSUHVHQWWKHQHZGHILQLWLRQIRURXWOLHUFOXVWHUEDVHGORFDORXWOLHU ZKLFKLV
PHDQLQJIXO DQG SURYLGHV LPSRUWDQFH WR WKH ORFDO GDWD EHKDYLRU $ PHDVXUH IRU LGHQWLI\LQJ WKH
SK\VLFDO VLJQLILFDQFH RI DQ RXWOLHU LV GHVLJQHG ZKLFK LV FDOOHG &%/2) &OXVWHU%DVHG /RFDO
2XWOLHU )DFWRU  :H DOVR SURSRVH WKH )LQG&%/2) DOJRULWKP IRU GLVFRYHULQJ RXWOLHUV 7KH
H[SHULPHQWDO UHVXOWV VKRZ WKDW RXU DSSURDFK RXWSHUIRUPHG WKH H[LVWLQJ PHWKRGV RQ LGHQWLI\LQJ
PHDQLQJIXODQGLQWHUHVWLQJRXWOLHUV 

.H\ZRUGV 2XWOLHU'HWHFWLRQ&OXVWHULQJ'DWD0LQLQJ


 ,QWURGXFWLRQ
$QRXWOLHULQDGDWDVHWLVGHILQHGLQIRUPDOO\DVDQREVHUYDWLRQWKDWLVFRQVLGHUDEO\GLIIHUHQW
IURP WKH UHPDLQGHUV DV LI LW LV JHQHUDWHG E\ D GLIIHUHQW PHFKDQLVP 0LQLQJ IRU RXWOLHUV LV DQ
LPSRUWDQWGDWDPLQLQJUHVHDUFKZLWKQXPHURXVDSSOLFDWLRQVLQFOXGLQJFUHGLWFDUGIUDXGGHWHFWLRQ
GLVFRYHU\ RI FULPLQDO DFWLYLWLHV LQ HOHFWURQLF FRPPHUFH ZHDWKHU SUHGLFWLRQ PDUNHWLQJ DQG
FXVWRPHUVHJPHQWDWLRQ
5HFHQWO\ VRPH VWXGLHV KDYH EHHQ SURSRVHG RQ RXWOLHU GHWHFWLRQ HJ .QRUU DQG 1J 
5DPDVZDP\ HW DO  %UHXQLJ HW DO$JJDUZDO DQG<X   IURP WKH GDWD PLQLQJ
FRPPXQLW\7KLV SDSHU SUHVHQWV D QHZ GHILQLWLRQ IRURXWOLHU QDPHO\ FOXVWHUEDVHG ORFDO RXWOLHU
ZKLFKLVLQWXLWLYHDQGPHDQLQJIXO7KLVZRUNLVPRWLYDWHGE\WKHIROORZLQJREVHUYDWLRQV
)LUVWO\DOOH[LVWLQJDOJRULWKPVIRURXWOLHUGHWHFWLRQLQYROYHKLJKFRPSXWDWLRQFRVWVZKLFKLV
QRW IHDVLEOH LQ WKH HQYLURQPHQW RI DFFHVV WR ODUJH GDWD VHWV VWRUHG LQ VHFRQGDU\ PHPRU\
)XUWKHUPRUHDOJRULWKPVSUHVHQWHGE\.QRUUDQG1J  5DPDVZDP\HWDO  %UHXQLJHW
DO  HWFGHILQHRXWOLHUVE\XVLQJWKHIXOOGLPHQVLRQDOGLVWDQFHVRIWKHSRLQWVIURPRQHDQG
DQRWKHU ZKLFK ZLOO UHVXOW LQ XQH[SHFWHG SHUIRUPDQFH DQG TXDOLWDWLYH FRVWV GXH WR WKH FXUVH RI
GLPHQVLRQDOLW\ 
6HFRQGO\ FOXVWHULQJ DOJRULWKPV OLNH 52&. *XKD HW DO   &3 1DQRSRXORV HW DO
 '%6&$1 (VWHU HWDO   FDQDOVR KDQGOH RXWOLHUV EXW WKHLU PDLQ FRQFHUQ LV WR ILQG
FOXVWHUVWKHRXWOLHUVDUHRIWHQUHJDUGHGDVQRLVHV$QGWKHUHDUHWKHIROORZLQJVKRUWFRPLQJVLQWKH
LQLWLDO ZRUN 6X HW DO   WKDW DGGUHVVHGFOXVWHULQJEDVHGRXWOLHU GHWHFWLRQ6XHW DO  
RQO\ UHJDUGHG VPDOO FOXVWHUV DV RXWOLHUV DQG D PHDVXUH IRU LGHQWLI\LQJ WKH GHJUHH RI HDFK REMHFW
EHLQJ DQ RXWOLHU LV QRW SUHVHQWHG )RU PRVW RI WKH GDWD SRLQWV LQ D GDWDVHW DUH QRW RXWOLHUV LW LV
PHDQLQJIXOWRLGHQWLI\RQO\WRSQRXWOLHUV,QWKLVFDVHWKHPHWKRGSURSRVHGE\6XHWDO  
IDLOHGWRIXOILOOWKLVWDVNHIIHFWLYHO\
)LQDOO\ XVLQJ WKH VDPH SURFHVV DQG IXQFWLRQDOLW\ WR VROYH ERWK FOXVWHULQJ DQG RXWOLHU
GLVFRYHU\LVKLJKO\GHVLUHG6XFKLQWHJUDWLRQZLOOEHDJUHDWEHQHILWWREXVLQHVVXVHUVEHFDXVHWKH\

&RUUHVSRQGLQJDXWKRU7HO H[W (PDLODGGUHVV ]HQJ\RXKH#\DKRRFRP =+H 

GR QRW QHHG WR ZRUU\ DERXW WKH VHOHFWLRQ RI GLIIHUHQW GDWD PLQLQJ DOJRULWKPV ,QVWHDG WKH\ FDQ
IRFXVRQGDWDDQGEXVLQHVVVROXWLRQ
%DVHG RQ WKH DERYH REVHUYDWLRQV ZH SUHVHQW WKH QHZ GHILQLWLRQ IRU RXWOLHU FOXVWHUEDVHG
ORFDO RXWOLHU $ PHDVXUH IRU LGHQWLI\LQJ WKH SK\VLFDO VLJQLILFDQFH RI DQ RXWOLHU QDPHO\ &%/2)
&OXVWHU%DVHG/RFDO2XWOLHU)DFWRU  LVDOVRGHILQHG)LQDOO\DIDVWDOJRULWKPIRUPLQLQJRXWOLHUV
LVSUHVHQWHGZKRVHHIIHFWLYHQHVVLVYHULILHGE\WKHH[SHULPHQWDOUHVXOWV
&RQWULEXWLRQVRIWKLVSDSHUDUHDVIROORZV
z :H SURSRVH D QRYHO GHILQLWLRQ IRU RXWOLHU FOXVWHUEDVHG ORFDO RXWOLHU ZKLFK KDV JUHDW
QHZLQWXLWLYHDSSHDODQGQXPHURXVDSSOLFDWLRQV
z $ PHDVXUHIRULGHQWLI\LQJWKHGHJUHHRIHDFKREMHFWEHLQJDQRXWOLHULVSUHVHQWHGZKLFK
LVFDOOHGFOXVWHUEDVHGORFDORXWOLHUIDFWRU
z :HSUHVHQWDQHIILFLHQWDOJRULWKPIRUPLQLQJFOXVWHUEDVHGORFDORXWOLHUVEDVHGRQRXU
GHILQLWLRQV 

7KHUHPDLQGHURIWKLVSDSHULV RUJDQL]HGDVIROORZV6HFWLRQGLVFXVVHVSUHYLRXVZRUN,Q
6HFWLRQ  ZH IRUPDOL]H RXU GHILQLWLRQ RI FOXVWHUEDVHG ORFDO RXWOLHU 6HFWLRQ  SUHVHQWV WKH
DOJRULWKPIRU PLQLQJGHILQHGRXWOLHUV([SHULPHQWDOUHVXOWVDUHJLYHQLQ6HFWLRQDQG6HFWLRQ
FRQFOXGHVWKHSDSHU

5HODWHG:RUN
7KHVWDWLVWLFVFRPPXQLW\FRQGXFWHGPRVWRIWKHSUHYLRXVVWXGLHVRQRXWOLHUPLQLQJ %DUQHWW
DQG/HZLV 7KHVHVWXGLHVFDQEHEURDGO\FODVVLILHGLQWRWZRFDWHJRULHV7KHILUVWFDWHJRU\LV
GLVWULEXWLRQEDVHG ZKHUH D VWDQGDUG GLVWULEXWLRQ LV XVHG WR ILW WKH GDWDVHW 2XWOLHUV DUH GHILQHG
EDVHGRQWKHSUREDELOLW\GLVWULEXWLRQ<DPDQLVKL7DNHXFKLDQG:LOOLDPV  XVHGD*DXVVLDQ
PL[WXUH PRGHO WRSUHVHQWWKHQRUPDOEHKDYLRUVDQGHDFKGDWXPLVJLYHQDVFRUHRQWKHEDVLVRI
FKDQJHVLQWKHPRGHO+LJKVFRUHLQGLFDWHVKLJKSRVVLELOLW\RIEHLQJDQRXWOLHU7KLVDSSURDFKKDV
EHHQ FRPELQHG ZLWK D VXSHUYLVHGEDVHG OHDUQLQJ DSSURDFK WR REWDLQ JHQHUDO SDWWHUQV IRU RXWOLHU
<DPDQLVKLDQG7DNHXFKL 7KHPDLQSUREOHPZLWKWKLVPHWKRGLVWKDWLWDVVXPHVWKDWWKH
XQGHUO\LQJGDWDGLVWULEXWLRQLVNQRZQDSULRU+RZHYHUIRUPDQ\DSSOLFDWLRQVLWLVDQLPSUDFWLFDO
DVVXPSWLRQ$QGWKHFRVWIRUILWWLQJGDWDZLWKVWDQGDUGGLVWULEXWLRQLVVLJQLILFDQWO\FRQVLGHUDEOH
'HSWKEDVHG LV WKH VHFRQG FDWHJRU\ IRU RXWOLHU PLQLQJ LQ VWDWLVWLFV 1XWV DQG 5RXVVHHXZ
 %DVHGRQVRPHGHILQLWLRQRIGHSWKGDWDREMHFWVDUHRUJDQL]HGLQFRQYH[KXOOOD\HUVLQGDWD
VSDFH DFFRUGLQJ WR SHHOLQJ GHSWK DQG RXWOLHUV DUH H[SHFWHG WR EH IRXQG IURP GDWD REMHFWV ZLWK
VKDOORZGHSWKYDOXHV,QWKHRU\GHSWKEDVHGPHWKRGVFRXOGZRUNLQKLJKGLPHQVLRQDOGDWDVSDFH
+RZHYHU GXHWRUHO\LQJ RQWKHFRPSXWDWLRQRINGFRQYH[KXOOVWKHVHWHFKQLTXHVKDYHDORZHU
ERXQGFRPSOH[LW\RI : 1 ZKHUH1 LVQXPEHURIGDWDREMHFWVDQGN LVWKHGLPHQVLRQDOLW\RI
N

WKHGDWDVHW7KLVPDNHVWKHVHWHFKQLTXHVLQIHDVLEOHIRUODUJHGDWDVHWZLWKKLJKGLPHQVLRQV
'LVWDQFHEDVHG RXWOLHU LV SUHVHQWHG E\ .QRUU DQG 1J   $ GLVWDQFHEDVHG RXWOLHU LQ D
GDWDVHW' LVDGDWDREMHFWZLWKSFW RIWKHREMHFWVLQ' KDYLQJDGLVWDQFHRIPRUHWKDQG DZD\
IURPLW7KLVQRWLRQJHQHUDOL]HVPDQ\FRQFHSWVIURPGLVWULEXWLRQEDVHGDSSURDFKDQGHQMR\VEHWWHU
PLQ

FRPSXWDWLRQDO FRPSOH[LW\ ,W LV IXUWKHU H[WHQGHG EDVHG RQ WKH GLVWDQFH RI D SRLQW IURP LWV N WK

QHDUHVWQHLJKERU 5DPDVZDP\HWDO $IWHUUDQNLQJSRLQWVE\WKHGLVWDQFHWRLWVN QHDUHVW WK

QHLJKERUWKHWRSNSRLQWVDUHLGHQWLILHGDVRXWOLHUV(IILFLHQWDOJRULWKPVIRUPLQLQJWRSNRXWOLHUV
DUH JLYHQ $OWHUQDWLYHO\ LQ WKH DOJRULWKP SURSRVHG E\ $QJLXOOL DQG 3L]]XWL   WKH RXWOLHU
IDFWRU RI HDFK GDWD SRLQW LV FRPSXWHG DV WKH VXP RI GLVWDQFHV IURP LWV N QHDUHVW QHLJKERUV 7KH


DERYHWKUHHDOJRULWKPVGHILQHRXWOLHUVE\XVLQJWKHIXOOGLPHQVLRQDOGLVWDQFHVRIWKHSRLQWVIURP
RQH DQRWKHU+RZHYHUUHFHQWUHVHDUFKUHVXOWVVKRZWKDWLQKLJKGLPHQVLRQDOVSDFHWKHFRQFHSWRI
SUR[LPLW\ PD\ QRW EH TXDOLWDWLYHO\ PHDQLQJIXO %H\HU HW DO   7KHUHIRUH WKH GLUHFW
DSSOLFDWLRQ RI GLVWDQFHEDVHG PHWKRGVWRKLJKGLPHQVLRQDOSUREOHPVRIWHQUHVXOWVLQXQH[SHFWHG
SHUIRUPDQFHDQGTXDOLWDWLYHFRVWVGXHWRWKHFXUVHRIGLPHQVLRQDOLW\ 
'HYLDWLRQEDVHGWHFKQLTXHVLGHQWLI\RXWOLHUVE\LQVSHFWLQJWKHFKDUDFWHULVWLFVRIREMHFWVDQG
FRQVLGHUDQREMHFWWKDWGHYLDWHVWKHVHIHDWXUHVDVDQRXWOLHU $UQLQJHWDO 
%UHXQLJ HW DO   LQWURGXFHG WKH FRQFHSW RI ³ORFDO RXWOLHU´ 7KH RXWOLHU UDQN RI D GDWD
REMHFWLVGHWHUPLQHGE\WDNLQJLQWRDFFRXQWWKHFOXVWHULQJVWUXFWXUHLQDERXQGHGQHLJKERUKRRGRI
WKHREMHFWZKLFKLVIRUPDOO\GHILQHGDV³ORFDORXWOLHUIDFWRU´ /2) 7KHLUQRWLRQVRIRXWOLHUVDUH
EDVHGRQWKHVDPHWKHRUHWLFDOIRXQGDWLRQRIGHQVLW\EDVHGFOXVWHULQJ (VWHUHWDO )RUWKH
FRPSXWDWLRQ RI ³GHQVLW\´ LV UHO\LQJ RQ IXOO GLPHQVLRQDO GLVWDQFH EHWZHHQ REMHFWV LQ KLJK
GLPHQVLRQDOVSDFHSUREOHPH[LVWVLQGLVWDQFHEDVHGPHWKRGVZLOOEHHQFRXQWHUHGDJDLQ
&OXVWHULQJ DOJRULWKPV OLNH 52&. *XKD HW DO   & 3 1DQRSRXORV HW DO  


'%6&$1 (VWHUHWDO FDQDOVRKDQGOHRXWOLHUVEXWWKHLUPDLQFRQFHUQLVWRILQGFOXVWHUV

WKH RXWOLHUV LQ WKH FRQWH[W RI FOXVWHULQJ DUH RIWHQ UHJDUGHG DV QRLVH ,Q JHQHUDO RXWOLHUV DUH
W\SLFDOO\MXVWLJQRUHGRUWROHUDWHGLQWKHFOXVWHULQJSURFHVVIRUWKHVHDOJRULWKPVDUHRSWLPL]HGIRU
SURGXFLQJPHDQLQJIXOFOXVWHUVZKLFKSUHYHQWVJLYLQJJRRGUHVXOWVRQRXWOLHUGHWHFWLRQ$QGWKHUH
DUHWKHIROORZLQJVKRUWFRPLQJVLQWKHLQLWLDOZRUN 6XHWDO<XHWDO WKDWDGGUHVVHG
FOXVWHULQJEDVHGRXWOLHUGHWHFWLRQ6XHWDO  RQO\UHJDUGHGVPDOOFOXVWHUVDVRXWOLHUVDQGD
PHDVXUHIRULGHQWLI\LQJWKHGHJUHHRIHDFKREMHFWEHLQJDQRXWOLHULVQRWSUHVHQWHG)RUPRVWRIWKH
GDWDSRLQWVLQDGDWDVHWDUHQRWRXWOLHUVLWLVPHDQLQJIXOWRLGHQWLI\RQO\WRSQRXWOLHUV+HQFHWKH
PHWKRG SURSRVHG E\ 6X HW DO   IDLOHG WR IXOILOO WKLV WDVN HIIHFWLYHO\ )XUWKHUPRUH KRZ WR
GLVWLQJXLVK VPDOO FOXVWHUV IRUP WKH UHVW LV QRW DGGUHVVHG LQ WKHLU PHWKRG <X HW DO  
LQWURGXFHG )LQG2XW D PHWKRG EDVHG RQ ZDYHOHW WUDQVIRUP WKDW LGHQWLILHV RXWOLHUV E\ UHPRYLQJ
FOXVWHUVIURPWKHRULJLQDOGDWDVHW 
$JJDUZDO DQG <X   GLVFXVVHG D QHZ WHFKQLTXH IRU RXWOLHU GHWHFWLRQ ZKLFK ILQGV
RXWOLHUVE\REVHUYLQJWKHGHQVLW\GLVWULEXWLRQRISURMHFWLRQVIURPWKHGDWD7KDWLVWKHLUGHILQLWLRQ
FRQVLGHUVDSRLQWWREHDQRXWOLHULILQVRPHORZHUGLPHQVLRQDOSURMHFWLRQLWLVSUHVHQWLQDORFDO
UHJLRQRIDEQRUPDOO\ORZGHQVLW\
7KHUHSOLFDWRUQHXWUDOQHWZRUN 511 LVHPSOR\HGWRGHWHFWRXWOLHUVE\+DUNLQVHWDO  
7KHDSSURDFKLVEDVHGRQWKHREVHUYDWLRQWKDWWKHWUDLQHGQHXWUDOQHWZRUNZLOOUHFRQVWUXFWVRPH
VPDOO QXPEHU RI LQGLYLGXDOV SRRUO\ DQG WKHVH LQGLYLGXDOV FDQ EH FRQVLGHUHG DV RXWOLHUV 7KH
RXWOLHUIDFWRUIRUUDQNLQJGDWDLVPHDVXUHGDFFRUGLQJWRWKHPDJQLWXGHRIWKHUHFRQVWUXFWLRQHUURU
$QLQWHUHVWLQJUHFHQWWHFKQLTXHILQGVRXWOLHUVE\LQFRUSRUDWLQJVHPDQWLFNQRZOHGJHVXFKDV
WKHFODVVODEHOV RI HDFKGDWDSRLQW LQ WKHGDWDVHW +H'HQJDQG;X ,QYLHZRIWKHFODVV
LQIRUPDWLRQDVHPDQWLFRXWOLHULVDGDWDSRLQWZKLFKEHKDYHVGLIIHUHQWO\ZLWKRWKHUGDWDSRLQWVLQ
WKHVDPHFODVV

&OXVWHU%DVHG/RFDO2XWOLHU
,Q WKLV VHFWLRQ ZH SURSRVH D QHZ GHILQLWLRQ IRU RXWOLHU FOXVWHUEDVHG ORFDO RXWOLHU %HIRUH
IRUPDOL]LQJWKHQHZGHILQLWLRQIRURXWOLHUVZHILUVWJLYHDQH[DPSOHWRLOOXVWUDWHRXUEDVLFLGHDV
&RQVLGHULQJ WKH G GDWD VHW LQ )LJXUH  7KHUH DUH  FOXVWHUV LQ WKLV ILJXUH &  &  & DQG & 
   
2EYLRXVO\ WKH GDWD SRLQWV LQ ERWK & DQG & VKRXOG EH UHJDUGHG DV RXWOLHUV DQG FDSWXUHG E\
SURSRVHG GHILQLWLRQV ,QWXLWLYHO\ ZH FDOO GDWD SRLQWV LQ & DQG & RXWOLHUV EHFDXVH WKH\ GLG QRW
 

EHORQJHWRWKHFOXVWHU& DQG&  7KXVLWLVUHDVRQDEOHWRGHILQHWKHRXWOLHUVIURPWKHSRLQWRIYLHZ

 

RI FOXVWHUVDQGLGHQWLI\WKRVHGDWDSRLQWVWKDWGRQ¶WOLHLQDQ\ODUJHFOXVWHUVDVRXWOLHUV+HUHWKH
 

QXPEHURIGDWDSRLQWVLQ& DQG& DUHGRPLQDQWLQWKHGDWDVHW)XUWKHUPRUHWRFDSWXUHWKHVSLULW

RI ³ORFDO´ SURSRVHGE\%UHXQLJHWDO  WKHFOXVWHUEDVHGRXWOLHUVVKRXOGVDWLVI\WKDWWKH\DUH
 

³ORFDO´WRVSHFLILHGFOXVWHUV)RUH[DPSOHGDWDSRLQWVLQ& DUH³ORFDO´ WR& 

7RLGHQWLI\WKHSK\VLFDOVLJQLILFDQFHRIWKHGHILQLWLRQRIDQRXWOLHUZHDVVLJQWRHDFKREMHFW
 

DQ RXWOLHU IDFWRU QDPHO\ &%/2) &OXVWHU%DVHG /RFDO 2XWOLHU )DFWRU  ZKLFK LV PHDVXUHG E\
ERWKWKHVL]HRIWKHFOXVWHUWKHREMHFWEHORQJVWRDQGWKHGLVWDQFHEHWZHHQWKHREMHFWDQGLWVFORVHVW
FOXVWHU LIWKHREMHFWOLHVLQDVPDOOFOXVWHU 


)LJGGDWDVHW'6

%HIRUH ZH SUHVHQW WKH FRQFHSW RI FOXVWHUEDVHG ORFDO RXWOLHUV DQG GHVLJQ WKH PHDVXUH IRU
RXWOLHUIDFWRUOHW¶VILUVWORRNDWWKHFRQFHSWDERXWFOXVWHULQJ7KURXJKRXWWKHSDSHUZHXVH_6_ WR
GRQDWHWKHVL]HRI6 ZKHUH6 LQJHQHUDODVHWFRQWDLQLQJVRPHHOHPHQWV

'HILQLWLRQ  /HW $  «$ EH D VHW RI DWWULEXWHV ZLWK GRPDLQV ' «' UHVSHFWLYHO\ /HW WKH
GDWDVHW ' EH D VHW RI UHFRUGV ZKHUH HDFK UHFRUG W W ' u « u '  7KH UHVXOWV RI D FOXVWHULQJ
 P  P

 ^&  &  « & ` ZKHUH & ˆ & I DQG &

 P

DOJRULWKP H[HFXWHG RQ ' LV GHQRWHG DV &   N L M 

‰ & ‰ « ‰ &
 N ' 7KHQXPEHURIFOXVWHULVN

+HUH WKHFOXVWHULQJ DOJRULWKP XVHG IRU SDUWLWLRQLQJ WKH GDWDVHW LQWR GLVMRLQW VHWV RI UHFRUGV
FDQEHFKRVHQIUHHO\7KHRQO\UHTXLUHPHQWIRUWKHVHOHFWHGFOXVWHULQJDOJRULWKPLVWKDWLWVKRXOG
KDYHWKHDELOLW\WRSURGXFHJRRGFOXVWHULQJUHVXOWV
$ FULWLFDOSUREOHPWKDWPXVWEHVROYHGEHIRUHGHILQLQJWKHFOXVWHUEDVHGORFDORXWOLHULVKRZ
WRLGHQWLI\ZKHWKHUDFOXVWHULVODUJHRUVPDOO 7KLVSUREOHPLVGLVFXVVHGLQGHILQLWLRQ

'HILQLWLRQ ODUJHDQGVPDOOFOXVWHU 6XSSRVH& ^&  &  «& ` LVWKHVHWRIFOXVWHUVLQWKH

VHTXHQFHWKDW_& _ t _& _ t « t _& _*LYHQWZRQXPHULFSDUDPHWHUV D DQG E  ZHGHILQHE DVWKH

  N

  N

ERXQGDU\RIODUJHDQGVPDOOFOXVWHULIRQHRIIROORZLQJIRUPXODVKROGV
 _& __& _«_& _ t _'_ D  

_t E
  E

_& __&
E E 

7KHQWKHVHWRIODUJHFOXVWHULVGHILQHGDV/& ^& _L d E`DQGWKHVHWRIVPDOOFOXVWHULVGHILQHG

DV6& ^& _M!E`
L

'HILQLWLRQJLYHVTXDQWLWDWLYHPHDVXUHWRGLVWLQJXLVKODUJHDQGVPDOOFOXVWHUV)RUPXOD  
FRQVLGHUVWKHIDFWWKDWPRVWGDWDSRLQWVLQWKHGDWDVHWDUHQRWRXWOLHUV7KHUHIRUHFOXVWHUVWKDWKROG
D ODUJHSRUWLRQRIGDWDSRLQWVVKRXOGEHWDNHQDVODUJHFOXVWHUV)RUH[DPSOHLI D LVVHWWRZH
LQWHQGWRUHJDUGFOXVWHUVFRQWDLQRIGDWDSRLQWVDVODUJHFOXVWHUV)RUPXOD  FRQVLGHUVWKH
IDFWWKDWODUJHDQGVPDOOFOXVWHUVVKRXOGKDYHVLJQLILFDQWGLIIHUHQFHVLQVL]H)RULQVWDQFHLWLVHDV\

WR JHW WKDW LI ZH VHW E WR  WKH VL]H RI DQ\ FOXVWHU LQ /& LV DW OHDVW  WLPHV RI WKH VL]H RI WKH

FOXVWHULQ6&

'HILQLWLRQ  &OXVWHU%DVHG /RFDO 2XWOLHU )DFWRU 6XSSRVH &  ^&  &  « & ` LV WKH VHW RI

FOXVWHUVLQWKHVHTXHQFHWKDW_& _ t _& _ t « t _& _ DQGWKHPHDQLQJVRI D  E  E  /&DQG6&DUH

  N

  N

WKHVDPHDVWKH\DUHIRUPDOL]HGLQ'HILQLWLRQ)RUDQ\UHFRUGW WKHFOXVWHUEDVHGORFDORXWOLHU
IDFWRURIWLVGHILQHGDV 
 _& _ PLQ GLVWDQFH W&  ZKHUHW  &  &  6&DQG&  /&IRUM WRE
&%/2) W                                                                   
L M L L M

 _& _  GLVWDQFH W& ZKHUHW  & DQG&  /&


L L L L

)URP'HILQLWLRQWKH&%/2)RIDUHFRUGLVGHWHUPLQHGE\WKHVL]HRILWVFOXVWHU DQGWKH
GLVWDQFH EHWZHHQ WKH UHFRUG DQG LWV FORVHVW FOXVWHU LI WKLV UHFRUG OLHV LQ VPDOO FOXVWHU  RU WKH
GLVWDQFHEHWZHHQWKHUHFRUGDQGWKHFOXVWHULWEHORQJVWR LIWKLVUHFRUGEHORQJVWRODUJHFOXVWHU 
ZKLFKSURYLGHVLPSRUWDQFHWRWKHORFDOGDWDEHKDYLRU 
)RUWKHFRPSXWDWLRQRIGLVWDQFHEHWZHHQWKHUHFRUGDQGWKHFOXVWHULWLVVXIILFLHQWWRDGRSW
WKHVLPLODULW\PHDVXUHXVHGLQWKHFOXVWHULQJDOJRULWKP

 $OJRULWKPIRU'HWHFWLQJ&OXVWHU%DVHG/RFDO2XWOLHUV

:LWKWKHRXWOLHUIDFWRU&%/2) ZHFDQGHWHUPLQHWKHGHJUHHRIDUHFRUG¶VGHYLDWLRQ,QWKLV
VHFWLRQZHZLOOGHVFULEHRXUDOJRULWKPIRUGHWHFWLQJRXWOLHUVDFFRUGLQJWR'HILQLWLRQ
 7R FRPSXWH &%/2) W  ZH QHHG D FOXVWHULQJ DOJRULWKP ILUVW ,Q WKLV SDSHU WKH FOXVWHULQJ
DOJRULWKP XVHG LV WKH 6TXHH]HU DOJRULWKP +H ;X DQG 'HQJ   ZKLFK FDQ SURGXFH JRRG
FOXVWHULQJUHVXOWVDQGDWWKHVDPHWLPHGHVHUYHVJRRGVDODELOLW\)RUWKHSURFHVVRIPLQLQJRXWOLHUV
LVWLJKWO\FRXSOHGZLWKWKHFOXVWHULQJDOJRULWKPZHJLYHDQLQWURGXFWLRQRQ6TXHH]HUDOJRULWKP 

7KH6TXHH]HU$OJRULWKP

/HW$  «$ EHDVHWRIFDWHJRULFDODWWULEXWHVZLWKGRPDLQV' «' UHVSHFWLYHO\/HW
WKHGDWDVHW' EH DVHWRIWXSOHVZKHUHHDFKWXSOHW W  ' u « u '  /HW7,'EHWKHVHWRIXQLTXH,'
 P  P

 P
RIHYHU\WXSOH)RUHDFKWLG  7,' WKHDWWULEXWHYDOXHIRU$ RIFRUUHVSRQGLQJWXSOHLVUHSUHVHQWHG
DVWLG$ 
L

'HILQLWLRQ &OXVWHU &OXVWHU ^WLG_WLG  7,'` LVVXEVHWRI7,'

L

'HILQLWLRQ*LYHQD&OXVWHU& WKHVHWRIGLIIHUHQWDWWULEXWHYDOXHVRQ$ ZLWKUHVSHFWWR& LV

GHILQHGDV9$/ & ^WLG$ _ WLG  &` ZKHUH d L d P
L

'HILQLWLRQ*LYHQD&OXVWHU& OHWD  '  WKHVXSSRUWRID LQ& ZLWKUHVSHFWWR$ LVGHILQHG

L L

DV6XS D _^WLG_WLG$ D `_

L L L L

'HILQLWLRQ 6XPPDU\ *LYHQD&OXVWHU& WKH6XPPDU\IRU& LVGHILQHGDV 

L L L

6XPPDU\ ^96 _  d L d P` ZKHUH96 ^ D  6XS D _D  9$/ & `

,QWXLWLYHO\ WKH 6XPPDU\ RI D &OXVWHU FRQWDLQV VXPPDU\ LQIRUPDWLRQ DERXW WKLV &OXVWHU ,Q
L L L L L L

JHQHUDOHDFK6XPPDU\ZLOOFRQVLVWVRIP HOHPHQWVZKHUHP LVQXPEHURIDWWULEXWHV7KHHOHPHQW

LQ6XPPDU\LVWKHVHWRISDLUVRIDWWULEXWHYDOXHVDQGWKHLUFRUUHVSRQGLQJVXSSRUWV 
'HILQLWLRQ &OXVWHU6WUXFWXUH&6 *LYHQD&OXVWHU& WKH&OXVWHU6WUXFWXUH &6 IRU& LV
GHILQHGDV&6 ^&OXVWHU6XPPDU\`
'HILQLWLRQ*LYHQD&OXVWHU& DQGDWXSOHW ZLWKWLG 7,' WKHVLPLODULW\EHWZHHQ& DQGWLG
LVGHILQHGDV 

¦
6XS DL
¦ 6XS D M
P
6LP &WLG  ZKHUHWLG$ L D
L

L


D 9$/ &

)URPWKHGHILQLWLRQLWLVFOHDUWKDWWKHVLPLODULW\XVHGKHUHLVVWDWLVWLFVEDVHG,QRWKHUZRUGV
M L

LI WKH VLPLODULW\ EHWZHHQ D WXSOH DQG DQ H[LVWHG FOXVWHU LV ODUJH HQRXJK LW PHDQV WKDW WKH
SUREDELOLW\RIWKHWXSOHEHORQJVWRWKLVFOXVWHULVODUJHU,QWKH6TXHH]HUDOJRULWKPWKLVPHDVXUHLV
XVHGWRGHWHUPLQHZKHWKHUWKHWXSOHVKRXOGEHSXWLQWRWKHFOXVWHURUQRW 
7KH6TXHH]HUDOJRULWKPKDVQWXSOHVDVLQSXWDQGSURGXFHFOXVWHUVDVILQDOUHVXOWV,QLWLDOO\
WKHILUVWWXSOHLQWKHGDWDEDVHLVUHDGLQDQGD&OXVWHU6WUXFWXUH &6 LVFRQVWUXFWHGZLWK& ^` 
7KHQWKHFRQVHTXHQWWXSOHVDUHUHDGLWHUDWLYHO\ 
)RU HYHU\ WXSOH E\ WKH VLPLODULW\ IXQFWLRQ ZH FRPSXWH LWV VLPLODULWLHV ZLWK DOO H[LVWLQJ
FOXVWHUV ZKLFK DUH UHSUHVHQWHG DQG HPERGLHG LQ WKH FRUUHVSRQGLQJ &6V 7KH ODUJHVW YDOXH RI
VLPLODULW\LVVHOHFWHGRXW,ILWLVODUJHUWKDQWKHJLYHQWKUHVKROGGRQDWHGDVV WKHWXSOHZLOOEHSXW
LQWRWKHFOXVWHUWKDWKDVWKHODUJHVWYDOXHRIVLPLODULW\7KH&6LVDOVRXSGDWHGZLWKWKHQHZWXSOH
,IWKHDERYHFRQGLWLRQGRHVQRWKROGDQHZFOXVWHUPXVWEHFUHDWHGZLWKWKLVWXSOH
7KHDOJRULWKPFRQWLQXHVXQWLOLWKDVWUDYHUVHGDOOWKHWXSOHVLQWKHGDWDVHW,WLVREYLRXVWKDW
WKH 6TXHH]HU DOJRULWKP RQO\ PDNHV RQH VFDQ RYHU WKH GDWDVHW WKXV KLJKO\ HIILFLHQW IRU GLVN
UHVLGHQWGDWDVHWVZKHUHWKH,2FRVWEHFRPHVWKHERWWOHQHFNRIHIILFLHQF\
7KH 6TXHH]HU DOJRULWKP LV SUHVHQWHG LQ )LJXUH  ,W DFFHSWV DV LQSXW WKH GDWDVHW ' DQG WKH
YDOXHRIWKHGHVLUHGVLPLODULW\WKUHVKROG7KHDOJRULWKPIHWFKHVWXSOHVIURP' LWHUDWLYHO\ 
,QLWLDOO\WKHILUVWWXSOHLVUHDGLQDQGWKHVXEIXQFWLRQDGG1HZ&OXVWHU6WUXFWXUH LVXVHGWR
HVWDEOLVKDQHZ&OXVWHULQJ6WUXFWXUH ZKLFKLQFOXGHV6XPPDU\DQG&OXVWHU 6WHS  
)RU WKH FRQVHTXHQW WXSOHV WKH VLPLODULW\ EHWZHHQ DQ H[LVWHG &OXVWHU & DQG HDFK WXSOH LV
FRPSXWHGXVLQJVXEIXQFWLRQVLP&RPSXWDWLRQ  6WHS :HJHWWKHPD[LPDOYDOXHRIVLPLODULW\
GRQDWHGE\VLPBPD[ DQGWKHFRUUHVSRQGLQJLQGH[RI&OXVWHU GRQDWHGE\LQGH[ IURPWKHDERYH
FRPSXWLQJ UHVXOWV 6WHS   7KHQ LI WKH VLPBPD[ LV ODUJHU WKDQ WKH LQSXW WKUHVKROG V
VXEIXQFWLRQ DGG7XSOH7R&OXVWHU  ZLOO EH FDOOHG WR DVVLJQ WKH WXSOH WR VHOHFWHG &OXVWHU 6WHS
 ,ILWLVQRWWKHFDVHWKHVXEIXQFWLRQDGG1HZ&OXVWHU6WUXFWXUH ZLOOEHFDOOHGWRFRQVWUXFW
D QHZ&6 6WHS )LQDOO\WKHFOXVWHULQJUHVXOWVZLOOEHODEHOHGRQWKHGLVN 6WHS 

$OJRULWKP6TXHH]HU ' V
%HJLQ
 ZKLOH ' KDVXQUHDGWXSOH ^
    WXSOH JHW&XUUHQW7XSOH '
    LI WXSOHWLG   ^
        DGG1HZ&OXVWHU6WUXFWXUH WXSOHWLG `
    HOVH^
       IRUHDFKH[LVWHGFOXVWHU&
  VLP&RPSXWDWLRQ &WXSOH  
       JHWWKHPD[YDOXHRIVLPLODULW\VLPBPD[
       JHWWKHFRUUHVSRQGLQJ&OXVWHU,QGH[LQGH[
       LIVLPBPD[! V
  DGG7XSOH7R&OXVWHU WXSOHLQGH[ 
       HOVH
          DGG1HZ&OXVWHU6WUXFWXUH WXSOHWLG `
  `
  RXWSXW&OXVWHULQJ5HVXOW 
(QG

)LJ6TXHH]HU$OJRULWKP

&KRRVLQJWKH6TXHH]HUDOJRULWKPDVWKHEDFNJURXQGFOXVWHULQJDOJRULWKPIRURXWOLHUGHWHFWLRQ
LQWKLVOLWHUDWXUHLVEDVHGRQWKHFRQVLGHUDWLRQWKDWWKLVDOJRULWKPKDVWKHIROORZLQJQRYHOIHDWXUHV
9 ,WDFKLHYHVERWKKLJKTXDOLW\RIFOXVWHULQJUHVXOWVDQGVFDODELOLW\
9 ,WVDELOLW\IRUKDQGOLQJKLJKGLPHQVLRQDOGDWDVHWVHIIHFWLYHO\
9 ,W GRHV QRW UHTXLUH WKH QXPEHU RI GHVLUHG FOXVWHUV DV DQ LQSXW SDUDPHWHU DQG LW FDQ
SURGXFH PRUH QDWXUDO FOXVWHUV ZLWK VLJQLILFDQW GLIIHUHQW VL]HV 7KLV IHDWXUH LV
XQGRXEWHGO\LPSRUWDQWIRUGLVFRYHULQJRXWOLHUVGHILQHGLQ6HFWLRQ  

7KH)LQG&%/2)$OJRULWKP

7KHDOJRULWKP)LQG&%/2)IRUGHWHFWLQJRXWOLHUVLVOLVWHGLQ)LJXUH
7KHDOJRULWKP)LQG&%/2)ILUVWSDUWLWLRQVWKHGDWDVHWLQWRFOXVWHUVZLWK6TXHH]HUDOJRULWKP
6WHS   7KH VHWV RI ODUJH FOXVWHUV DQG VPDOO FOXVWHUV /& DQG 6& DUH GHULYHG XVLQJ WKH
SDUDPHWHUVDFFRUGLQJWRGHILQLWLRQ 6WHS 7KHQIRUHYHU\GDWDSRLQWLQWKHGDWDVHWWKHYDOXH
RI &%/2)LVFRPSXWHGZLWKGHILQLWLRQ 6WHS  
 $OJRULWKP)LQG&%/2)

,QSXW   ' $  «$   WKHGDWDVHW

 7KHSDUDPHWHU D DQG  E SDUDPHWHUV
 P

 2XWSXW 7KHYDOXHVRI&%/2)IRUDOOUHFRUGV  LQGLFDWHVWKHGHJUHHRIGHYLDWLRQ


EHJLQ
 &OXVWHULQJWKHGDWDVHW' $  «$  XVLQJ6TXHH]HUDOJRULWKP
 P

  3URGXFHGFOXVWHUV& ^&  &  «& ` DQG_& _! _& _! «! _& _ 

  N   N

 *HW/&DQG6&ZLWKWKHSDUDPHWHUV  JHWWKHVHWRIODUJHDQGVPDOOFOXVWHUV

 IRUHDFKUHFRUGW LQWKHGDWDVHWGREHJLQ 
  LIW EHORQJVWR&L DQG&L EHORQJVWR6&GREHJLQ
  &%/2) _& _ PLQ GLVWDQFH W& 
L M
& EHORQJVWR/&
M

  HOVH 
  &%/2) _& _ GLVWDQFH W&L & EHORQJVWR/&

L L

 UHWXUQ&%/2)
 HQG
  HQG

)LJ7KH)LQG&%/2)$OJRULWKP

$OJRULWKP)LQG&%/2)KDVWZRSDUWV &OXVWHULQJWKHGDWDVHWDQG &RPSXWLQJWKHYDOXH

RI &%/2) IRU HDFK UHFRUG 7KH 6TXHH]HU GHWHUPLQHV WKH FRVW RI SDUW  )URP WKH H[HFXWLRQ
SURFHVVRI6TXHH]HU LWQHHGV RQO\RQHVFDQRYHUWKHGDWDVHW7KXVWKHFRVWRISDUWVKRXOGEH
2 1 ZKHUH1 LVQXPEHURIWKHUHFRUGVLQWKHGDWDVHW$VWRWKHSDUWRQHVFDQRYHUWKHGDWDVHWLV
DOVRUHTXLUHG7KHUHIRUHWKHRYHUDOOFRVWIRUWKH)LQG&%/2)DOJRULWKPLV2 1 
)URPWKHDERYHDQDO\VLVZHFDQVHHWKDWOLQHDUVFDODELOLW\LVDFKLHYHGZLWKUHVSHFWWRWKHVL]H
RI GDWDVHWZKLFKPDNHVWKH)LQG&%/2)DOJRULWKPTXDOLILHGIRUKDQGOLQJODUJHGDWDVHW


 ([SHULPHQWDO5HVXOWV

$ FRPSUHKHQVLYH SHUIRUPDQFH VWXG\ KDV EHHQ FRQGXFWHG WR HYDOXDWH RXU DOJRULWKP ,Q WKLV
VHFWLRQ ZH GHVFULEH WKRVH H[SHULPHQWV DQG WKHLU UHVXOWV :H UDQRXU DOJRULWKP RQ ERWK UHDOOLIH
GDWDVHWV REWDLQHG IURP WKH 8&, 0DFKLQH /HDUQLQJ 5HSRVLWRU\ 0HU] DQG 0HUSK\   DQG
V\QWKHWLFGDWDVHWV2XUDOJRULWKPZDVLPSOHPHQWHGLQ-DYD$OOH[SHULPHQWVZHUHFRQGXFWHGRQD
3HQWLXP PDFKLQHZLWK0RI5$0DQGUXQQLQJ:LQGRZV6HUYHU
 :H XVHG WKUHH UHDO OLIH GDWDVHWV WR GHPRQVWUDWH WKH HIIHFWLYHQHVV RI RXU DOJRULWKPV DJDLQVW
RWKHU DOJRULWKPV )RU DOO WKH H[SHULPHQWV WKH WZR SDUDPHWHUV QHHGHG E\ )LQG&%/2)DUH VHW WR
 DQG  VHSDUDWHO\ )RU WKH .11 DOJRULWKP 5DPDVZDP\ HW DO   WKH UHVXOWV ZHUH
REWDLQHGXVLQJWKHQHDUHVWQHLJKERXU WKHUHVXOWVGLGQ¶WFKDQJHVLJQLILFDQWO\ZKHQWKHSDUDPHWHU
N LVVSHFLILHGWRDOWHUQDWLYHYDOXHV6LQFHWKH6TXHH]HUDOJRULWKPRSHUDWHVRQFDWHJRULFDOGDWDVHW
WKH$QQHDOLQJGDWDVHWDUHGLVUHWL]HG XVLQJWKHDXWRPDWLFGLVFUHWL]DWLRQIXQFWLRQDOLW\ SURYLGHGE\
WKH&%$ /LXHWDO VRIWZDUH
 $QQHDOLQJ'DWD

7KHILUVWGDWDVHWXVHGLVWKH$QQHDOLQJGDWDVHWZKLFKKDVLQVWDQFHVZLWKDWWULEXWHV
7KHGDWDVHWFRQWDLQVDWRWDORI QRQHPSW\ FODVVHV&ODVVKDVWKHODUJHVWQXPEHURILQVWDQFHV
7KHUHPDLQHGFODVVHVDUHUHJDUGHGDVUDUHFODVVODEHOVIRUWKH\DUHVPDOOLQVL]H7KHFRUUHVSRQGLQJ
FODVVGLVWULEXWLRQLVLOOXVWUDWHGLQ7DEOH 

7DEOH&ODVV'LVWULEXWLRQRI$QQHDOLQJ'DWD6HW
&DVH &ODVVFRGHV 3HUFHQWDJHRILQVWDQFHV
&RPPRQO\2FFXUULQJ&ODVVHV  
5DUH&ODVVHV 8 

$VSRLQWHGRXWE\$JJDUZDODQG<X  RQHZD\WRWHVWKRZZHOOWKHRXWOLHUGHWHFWLRQ

DOJRULWKP ZRUNHG LV WR UXQ WKH PHWKRG RQ WKH GDWDVHW DQG WHVW WKH SHUFHQWDJH RI SRLQWV ZKLFK
EHORQJWRWKHUDUHFODVVHV,IRXWOLHUGHWHFWLRQZRUNVZHOOLWLVH[SHFWHGWKDWWKHUDUHFODVVHVZRXOG
EHRYHUUHSUHVHQWHGLQWKHVHWRISRLQWVIRXQG7KHVHNLQGVRIFODVVHVDUHDOVRLQWHUHVWLQJIURPD
SUDFWLFDOSHUVSHFWLYH 
7DEOHVKRZVWKHUHVXOWVSURGXFHGE\WKH)LQG&%/2)DOJRULWKPDJDLQVWWKH.11DOJRULWKP
5DPDVZDP\HWDO +HUHWKHWRSUDWLRLVUDWLRRIWKHQXPEHURIUHFRUGVVSHFLILHGDVWRSN
RXWOLHUVWRWKDWRIWKHUHFRUGVLQWKHGDWDVHW7KHFRYHUDJHLVUDWLRRIWKHQXPEHURIGHWHFWHGUDUH
FODVVHVWRWKDWRIWKHUDUHFODVVHVLQWKHGDWDVHW)RUH[DPSOHZHOHW)LQG&%/2)DOJRULWKPILQG
WKHWRSRXWOLHUVZLWKWKHWRSUDWLRRI%\H[DPLQLQJWKHVHSRLQWVZHIRXQGWKDW
RIWKHPEHORQJHGWRWKHUDUHFODVVHV,QFRQWUDVWZKHQZHUDQWKH.11DOJRULWKPRQWKLVGDWDVHW
ZHIRXQGWKDWRQO\RIWRSRXWOLHUVEHORQJHGWRUDUHFODVVHV 

7DEOH'HWHFWHG5DUH&ODVVHVLQ$QQHDOLQJ'DWDVHW
7RS5DWLR 1XPEHURI 1XPEHURI5DUH&ODVVHV,QFOXGHG &RYHUDJH 
5HFRUGV  )LQG&%/2) .11
        
        
        
        
        

)URP7DEOHWKHSHUIRUPDQFHRIWKH)LQG&%/2)DOJRULWKPRXWSHUIRUPHGWKDWRIWKH.11
DOJRULWKP LQ DOO WKH ILYH FDVHV HVSHFLDOO\ ZKHQ WKH WRS UDWLR LV UHODWLYH VPDOO WKH )LQG&%/2)
DOJRULWKPZRUNHGPXFKEHWWHU

 /\PSKRJUDSK\'DWD

7KH VHFRQG GDWDVHW XVHG LV WKH /\PSKRJUDSK\ GDWD VHW ZKLFK KDV  LQVWDQFHV ZLWK 
DWWULEXWHV7KHGDWDVHWFRQWDLQVDWRWDORIFODVVHV&ODVVHVDQGKDYHWKHODUJHVWQXPEHURI
LQVWDQFHV7KH UHPDLQHG FODVVHV DUH UHJDUGHG DV UDUH FODVV ODEHOV IRU WKH\ DUH VPDOO LQ VL]H7KH
FRUUHVSRQGLQJFODVVGLVWULEXWLRQLVLOOXVWUDWHGLQ7DEOH
 7DEOH&ODVV'LVWULEXWLRQRI/\PSKRJUDSK\'DWD6HW
&DVH &ODVVFRGHV 3HUFHQWDJHRILQVWDQFHV
&RPPRQO\2FFXUULQJ&ODVVHV  
5DUH&ODVVHV  

7DEOHVKRZVWKHUHVXOWVSURGXFHGE\WKH)LQG&%/2)DOJRULWKPDJDLQVWWKH.11DOJRULWKP
,QWKLVH[SHULPHQWWKH)LQG&%/2)DOJRULWKPFDQILQGDOOWKHUHFRUGVLQUDUHFODVVHVZKHQWKHWRS
UDWLRUHDFKHG0RUHRYHULWFDQILQGPDMRULW\RIWKHUHFRUGVLQWKHUDUHFODVVHVHYHQWKHWRS
UDWLRLVVHWWRUHODWLYHVPDOO,QFRQWUDVWWKHSHUIRUPDQFHRI.11DOJRULWKPLVQRWVDWLVILHG

7DEOH'HWHFWHG5DUH&ODVVHVLQ/\PSKRJUDSK\'DWDVHW
7RS5DWLR 1XPEHURI 1XPEHURI5DUH&ODVVHV,QFOXGHG &RYHUDJH 
5HFRUGV  )LQG&%/2) .11
        
        
        
        

 :LVFRQVLQ%UHDVW&DQFHU'DWD

7KHWKLUGGDWDVHWXVHGLVWKH:LVFRQVLQEUHDVWFDQFHUGDWDVHWZKLFKKDVLQVWDQFHVZLWK
DWWULEXWHV (DFK UHFRUG LV ODEHOHG DV EHQLJQ  RU   RU PDOLJQDQW  RU   :H
IROORZWKHH[SHULPHQWDOWHFKQLTXHRI+DUNLQVHWDO  E\UHPRYLQJVRPHRIWKHPDOLJQDQW
UHFRUGV WR IRUP D YHU\ XQEDODQFHG GLVWULEXWLRQ WKH UHVXOWDQW GDWDVHW KDG    PDOLJQDQW
UHFRUGVDQG  EHQLJQUHFRUGV
)RUWKLVGDWDVHWRXUDLPLVWRWHVWWKHSHUIRUPDQFHRIRXUDOJRULWKPZLWKWKH.11DOJRULWKP
DQGWKH511EDVHGRXWOLHUGHWHFWLRQDOJRULWKP +DUNLQVHWDO 7KHUHVXOWVRI511EDVHG
RXWOLHUGHWHFWLRQDOJRULWKPRQWKLVGDWDVHWDUHUHSRUWHGIURP+DUNLQVHWDO   
7DEOH  VKRZV WKH UHVXOWV SURGXFHG E\ WKH )LQG&%/2) DOJRULWKP DJDLQVW ERWK WKH .11
DOJRULWKPDQGWKH511EDVHGRXWOLHUGHWHFWLRQDOJRULWKP 
2QH LPSRUWDQW REVHUYDWLRQ IURP 7DEOH  ZDV WKDW RXU DOJRULWKP SHUIRUPHG WKH EHVW IRU DOO
FDVHVDQGQHYHUSHUIRUPHGWKHZRUVW7KDWLVWRVD\WKH)LQG&%/2)DOJRULWKPLVPRUHFDSDEOHWR
HIIHFWLYHO\GHWHFWRXWOLHUVWKDQWKHRWKHUWZRDOJRULWKPV
$QRWKHULPSRUWDQW REVHUYDWLRQ ZDVWKDWWKH)LQG&%/2)DOJRULWKPIRXQGDOOWKHPDOLJQDQW
UHFRUGVZLWK WKHWRSUDWLRDW ,QFRQWUDVWIRUWKH511EDVHGRXWOLHUGHWHFWLRQDOJRULWKPLW
DFKLHYHGWKLVJRDOZLWKWKHWRSUDWLRDWZKLFKLVDOPRVWWKHWZLFHIRUWKDWRI)LQG&%/2)
,QVXPPDU\WKHDERYHH[SHULPHQWDOUHVXOWVRQWKHWKUHHGDWDVHWVVKRZWKDWWKH)LQG&%/2)
DOJRULWKP FDQRXWOLHUV PRUHHIILFLHQWO\WKDQH[LVWLQJDOJRULWKPVIURPZKLFKZHFDQFRQILGHQWO\
DVVHUWWKDWWKHQHZFRQFHSWRIFOXVWHUEDVHGORFDORXWOLHULVSURPLVLQJLQSUDFWLFH



7KHUHVXOWDQWGDWDVHWLVSXEOLFDYDLODEOHDWKWWSUHVHDUFKFPLVFVLURDXURKDQERXWOLHUVEUHDVWFDQFHU
 7DEOH'HWHFWHG0DOLJQDQW5HFRUGVLQ:LVFRQVLQ%UHDVW&DQFHU'DWDVHW
7RS5DWLR 1XPEHU 1XPEHURI0DOLJQDQW,QFOXGHG &RYHUDJH 
RI5HFRUGV  )LQG&%/2) 511 .11
           
           
           
           
           
           
           
           
           
           
           
           
           
           

 6FDODELOLW\7HVWV

,Q WKLV VHFWLRQ ZH HYDOXDWH WKH SHUIRUPDQFH RI WKH FRPSXWDWLRQ RI &%/2) 7KH GDWDVHWV
ZHUH JHQHUDWHG XVLQJ D GDWD JHQHUDWRU LQ ZKLFK DOO SRVVLEOH YDOXHV DUH SURGXFHG ZLWK
DSSUR[LPDWHO\ HTXDOSUREDELOLW\DQGWKHQXPEHURIDWWULEXWHYDOXHVIRUHDFKDWWULEXWHVHWWR
7R ILQG RXW KRZ WKH QXPEHU RI UHFRUGV DQG DWWULEXWHV DIIHFW WKH DOJRULWKP ZH UDQ D VHULHV RI
H[SHULPHQWVZLWKLQFUHDVLQJQXPEHURIUHFRUGVDQGDWWULEXWHV7KHQXPEHURIUHFRUGVYDULHGIURP
 PLOOLRQ WR  PLOOLRQ DQG WKH QXPEHU RI DWWULEXWHV YDULHG IURP  WR 7KH HODSVHG WLPH
PHDVXUHGLVVKRZQLQ)LJXUH


F
H
V  G G
H
P G G
L
7
Q 

R
L
W 
X
F
[ 
H

(

    


7KHQXPEHURIWXSOHV LQPLOOLRQ

)LJ5XQWLPHIRUWKHFRPSXWDWLRQRI&%/2)VZLWKGLIIHUHQWGDWDVHWVL]HVDQGGLIIHUHQWQXPEHURIDWWULEXWHV  

)URPWKLVILJXUHZHFDQVHHWKDWOLQHDUVFDODELOLW\LVDFKLHYHGZLWKUHVSHFWWRWKHQXPEHURI
UHFRUGVDQGDWWULEXWHV7KLVSURSHUW\TXDOLILHVRXUDOJRULWKPIRUGLVFRYHULQJRXWOLHUVLQYHU\ODUJH
GDWDEDVHV   

 &RQFOXVLRQV

,Q WKLV SDSHU ZH SUHVHQW D QHZ GHILQLWLRQ IRU RXWOLHU FOXVWHUEDVHG ORFDO RXWOLHU ZKLFK LV
LQWXLWLYH DQG SURYLGHV LPSRUWDQFH WR WKH ORFDO GDWD EHKDYLRU $ PHDVXUH IRU LGHQWLI\LQJ WKH
SK\VLFDOVLJQLILFDQFHRIDQRXWOLHUQDPHO\&%/2) &OXVWHU%DVHG/RFDO2XWOLHU)DFWRU  LVDOVR
GHILQHG )XUWKHUPRUH ZH SURSRVH WKH )LQG&%/2) DOJRULWKP IRU GLVFRYHULQJ RXWOLHUV 7KH
H[SHULPHQWDO UHVXOWV VKRZ WKDW RXU DSSURDFK RXWSHUIRUPHG H[LVWLQJ PHWKRGV RQ LGHQWLI\LQJ
PHDQLQJIXODQGLQWHUHVWLQJRXWOLHUV 
)RU IXWXUH ZRUN ZH ZLOO LQWHJUDWH WKH )LQG&%/2) DOJRULWKP PRUH WLJKWO\ ZLWK FOXVWHULQJ
DOJRULWKPVWRPDNHWKHGHWHFWLQJSURFHVVPRUHHIILFLHQW7KHGHVLJQLQJRIHIIHFWLYHWRSNRXWOLHUV¶
GHWHFWLRQDOJRULWKPZLOOEHDOVRDGGUHVVHG


$FNQRZOHGJHPHQWV 

7KH FRPPHQWV DQG VXJJHVWLRQV IURP WKH DQRQ\PRXV UHYLHZHUV JUHDWO\ LPSURYH WKLV SDSHU 7KH
DXWKRUV ZLVK WR WKDQN )DEUL]LR $QJLXOOL IRU VHQGLQJ XV WKHLU 3.''¶ SDSHU 6SHFLDO
DFNQRZOHGJHPHQW JRHV WR 0EDOH -DPHVRQ IRU KLV KHOS RQ (QJOLVK HGLWLQJ 7KH 1DWLRQDO 1DWXUH
6FLHQFH)RXQGDWLRQRI35&KLQD 1R DQGWKH,%06855HVHDUFK)XQGVXSSRUWHG
WKLVUHVHDUFK


5HIHUHQFHV
$JJDUZDO &<X 3 2XWOLHUGHWHFWLRQIRUKLJKGLPHQVLRQDOGDWD,Q3URFHHGLQJVRI6,*02'¶6DQWD
%DUEDUD&$86$SS
$QJLXOOL)3L]]XWL&)DVWRXWOLHUGHWHFWLRQLQKLJKGLPHQVLRQDOVSDFHV,Q3URFHHGLQJVRI3.''¶
$UQLQJ $ $JUDZDO 5 5DJKDYDQ 3  $ OLQHDU PHWKRG IRU GHYLDWLRQ GHWHFWLRQ LQ ODUJH GDWDEDVHV ,Q
3URFHHGLQJVRI.''¶3RUWODQG2586$SS
%DUQHWW9/HZLV72XWOLHUVLQVWDWLVWLFDOGDWD-RKQ:LOH\DQG6RQV1HZ<RUN
%H\HU . *ROGVWHLQ - 5DPDNULVKQDQ 5 6KDIW 8  :KHQ LV ³QHDUHVW QHLJKERUV´ PHDQLQJIXO" ,Q
3URFHHGLQJVRI,&'7¶-HUXVDOHP,VUDHOSS
%UHXQLJ 0 0 .ULHJHO + 3 1J 5 7 6DQGHU -  /2) LGHQWLI\LQJ GHQVLW\EDVHG ORFDO RXWOLHUV ,Q
3URFHHGLQJVRI6,*02'¶'DOODV7H[DVSS
(VWHU 0 .ULHJHO + 3 6DQGHU - ;X ;  $ GHQVLW\EDVHG DOJRULWKP IRU GLVFRYHULQJ FOXVWHUV LQ ODUJH
VSDWLDOGDWDEDVHV´,Q3URFHHGLQJVRI.''¶3RUWODQG2586$SS
*XKD 6 5DVWRJL 5 .\XVHRN 6  52&. $ UREXVW FOXVWHULQJ DOJRULWKP IRU FDWHJRULFDO DWWULEXWHV ,Q
3URFHHGLQJVRI,&'(¶6\GQH\$XVWUDOLDSS
+DUNLQV6+H+:LOODPV*-%DVWHU5$2XWOLHUGHWHFWLRQXVLQJUHSOLFDWRUQHXUDOQHWZRUNV,Q3URF
RI WKH WK ,QWHUQDWLRQDO &RQIHUHQFH RQ 'DWD :DUHKRXVLQJ DQG .QRZOHGJH 'LVFRYHU\ $L[HQ3URYHQFH
)UDQFHSS
+H='HQJ6;X;2XWOLHUGHWHFWLRQLQWHJUDWLQJVHPDQWLFNQRZOHGJH,Q 3URFRIWKHWK,QW¶O&RQI
RQ:HE$JH,QIRUPDWLRQ0DQDJHPHQW %HLMLQJ&KLQDSS
+H = ;X ; 'HQJ 6  6TXHH]HU $Q HIILFLHQW DOJRULWKP IRU FOXVWHULQJ FDWHJRULFDO GDWD -RXUQDO RI
 &RPSXWHU6FLHQFHDQG7HFKQRORJ\  
-LDQJ 0 ) 7VHQJ 6 6 6X & 0  7ZRSKDVH FOXVWHULQJ SURFHVV IRU RXWOLHUV GHWHFWLRQ 3DWWHUQ
5HFRJQLWLRQ/HWWHUV  
.QRUU(01J57$OJRULWKPVIRUPLQLQJGLVWDQFHEDVHGRXWOLHUVLQODUJHGDWDVHWV  ,Q3URFHHGLQJVRI
9/'%¶1HZ<RUN86$SS
/LX%+VX:0D<,QWHJUDWLQJFODVVLILFDWLRQDQGDVVRFLDWLRQUXOHPLQLQJ,Q3URFHHGLQJVRI.''¶
1HZ<RUN86$SS
0HU] & - 0HUSK\ 3  8&, 5HSRVLWRU\ RI 0DFKLQH /HDUQLQJ 'DWDEDVHV 85/
+WWSZZZLFVXFLHGXaPOHDUQ0/55HSRVLWRU\KWPO
1DQRSRXORV$7KHRGRULGLV<0DQRORSRXORV<&3&OXVWHULQJEDVHGRQFORVHVWSDLUV,Q3URFHHGLQJV
RI9/'%¶5RPH,WDO\SS
1XWV 5 5RXVVHHXZ 3  &RPSXWLQJ GHSWK FRQWRXUV RI ELYDULDWH SRLQW FORXGV  -RXUQDO RI &RPSXWDWLRQDO
6WDWLVWLFVDQG'DWD$QDO\VLVYRO
5DPDVZDP\65DVWRJL5.\XVHRN6(IILFLHQWDOJRULWKPVIRUPLQLQJRXWOLHUVIURPODUJHGDWDVHWV,Q
3URFHHGLQJVRI6,*02'¶'DOODV7H[DVSS
<DPDQLVKL.7DNHXFKL-'LVFRYHULQJRXWOLHUILOWHULQJUXOHVIURPXQODEHOHGGDWDFRPELQLQJDVXSHUYLVHG
OHDUQHUZLWKDQXQVXSHUYLVHG/HDUQHU,Q3URFHHGLQJVRI.''¶SS
<DPDQLVKL.7DNHXFKL-:LOOLDPV*2QOLQHXQVXSHUYLVHGRXWOLHUGHWHFWLRQXVLQJILQLWHPL[WXUHVZLWK
GLVFRXQWLQJOHDUQLQJDOJRULWKPV,Q3URFHHGLQJVRI.''¶%RVWRQ0$86$SS
<X ' 6KHLNKROHVODPL * =KDQJ $  )LQG2XW )LQGLQJ RXW RXWOLHUV LQ ODUJH GDWDVHWV 7HFKQLTXH 5HSRUW
6WDWH8QLYHUVLW\RI1HZ<RUNDW%XIIDOR