• No results found

Human demographic history : Insights on the human past based on genomes from Southern through Central Africa

N/A
N/A
Protected

Academic year: 2021

Share "Human demographic history : Insights on the human past based on genomes from Southern through Central Africa"

Copied!
114
0
0

Loading.... (view fulltext now)

Full text

(1)

ACTA UNIVERSITATIS

UPSALIENSIS UPPSALA

2020

Digital Comprehensive Summaries of Uppsala Dissertations

from the Faculty of Science and Technology

1949

Human demographic history

Insights on the human past based on genomes from

Southern through Central Africa

GWENNA BRETON

ISSN 1651-6214 ISBN 978-91-513-0979-8 urn:nbn:se:uu:diva-416653

(2)

Dissertation presented at Uppsala University to be publicly examined in Ekmansalen, EBC, Norbyvägen 14, Uppsala, Friday, 18 September 2020 at 13:15 for the degree of Doctor of Philosophy. The examination will be conducted in English. Faculty examiner: Professor David Comas (Universitat Pompeu Fabra, Barcelona, Spain).

Abstract

Breton, G. 2020. Human demographic history. Insights on the human past based on genomes

from Southern through Central Africa. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 1949. 112 pp. Uppsala: Acta

Universitatis Upsaliensis. ISBN 978-91-513-0979-8.

Evidence from paleontology, archaeology and population genetics support that modern humans originated in Africa. While the out-of-Africa event and subsequent colonization of all continents are well documented, human history in Africa at that time and before is less studied. Some current-day hunter-gatherer populations trace most of their genetic lineages to populations who inhabited Sub-Saharan Africa until the arrival of farming. They are informative about human history before and after the arrival of farming.

I studied high-coverage genomes from two such groups, the Khoe-San from Southern Africa and the rainforest hunter-gatherers from Central Africa. I generated a total of 74 genomes, significantly increasing the number of genomes from Sub-Saharan African hunter-gatherers. I compared several versions of a commonly used pipeline for high-coverage genomes and showed that using standard ascertained reference datasets has no significant impact on variant calling in populations from Sub-Saharan Africa. Using the full genome information, I described the genetic diversity in the Khoe-San and in the rainforest hunter-gatherers and showed that gene flow from agropastoralist groups increased the Khoe-San genetic diversity. I also detected a signal of population size decline in the Khoe-San around the time of the out-of-Africa event, and I evaluated the power of the method to detect bottlenecks by applying it to simulated data. I investigated the history of modern humans in Africa by estimating divergence times between populations and applying an Approximate Bayesian Computation analysis. We confirmed that the earliest divergence event was between the Khoe-San ancestral lineage and the rest of modern humans, ~250-350 kya. I also showed that the possibility of high gene flow should be incorporated in models of human evolution.

I furthermore examined SNP array data for two BaTwa populations from Zambia and showed that 20-30% of their autosomal diversity is hunter-gatherer-like. The estimated times for the admixture between a presumably local hunter-gatherer population and incoming agropastoralist groups are consistent with archaeological records.

In this thesis, I investigated questions related to human history in Sub-Saharan Africa, from the emergence of modern humans ~300 kya to recent events related to the expansion of farming.

Keywords: high-coverage genomes, Sub-Saharan Africa, demography, genetic admixture,

population genetics, human evolutionary genetics, Khoe-San, rainforest hunter-gatherers

Gwenna Breton, Department of Organismal Biology, Human Evolution, Norbyvägen 18 A, Uppsala University, SE-752 36 Uppsala, Sweden.

© Gwenna Breton 2020 ISSN 1651-6214 ISBN 978-91-513-0979-8

(3)

3RXU $PDQGLQH TX¶LO PH WDUGH GH UHQFRQWUHU HW SRXU PD 0DPLH TX¶LO PH WDUGH GH UHYRLU

(4)
(5)

/LVW RI SDSHUV

7KLV WKHVLV LV EDVHG RQ WKH IROORZLQJ SDSHUV ZKLFK DUH UHIHUUHG WR LQ WKH WH[W E\ WKHLU 5RPDQ QXPHUDOV

, 6FKOHEXVFK &0  6M|GLQ 3  %UHWRQ *  *QWKHU 7 1DLGRR 7 +ROOIHOGHU 1 6M|VWUDQG $( ;X - *DWWHSDLOOH /0 9LFHQWH 0 6FRILHOG '* 0DOPVWU|P + GH -RQJK 0 /RPEDUG 0 6RRG\DOO + -DNREVVRQ 0  .KRH6DQ *HQRPHV 5HYHDO 8QLTXH 9DULDWLRQ DQG &RQILUP WKH 'HHSHVW 3RSXODWLRQ 'LYHUJHQFH LQ +RPR VDSLHQV

0ROHFXODU %LRORJ\ DQG (YROXWLRQ LQ SUHVV

,, %UHWRQ * -RKDQVVRQ $ 6M|GLQ 3 6FKOHEXVFK &0 -DNREVVRQ 0  &RPSDULVRQ RI VHTXHQFLQJ GDWD SURFHVVLQJ SLSHOLQHV DQG

DSSOLFDWLRQ WR XQGHUUHSUHVHQWHG KXPDQ SRSXODWLRQV 8QGHU UHYLHZ ,,, %UHWRQ * 6M|GLQ 3 =HUYDNLV 3, /DXUHQW 5 6M|VWUDQG $(

+HZOHWW %6 %DUUHLUR /% 3HUU\ *+ 6RRG\DOO + +H\HU ( 6FKOHEXVFK &0 9HUGX 3 -DNREVVRQ 0  'HFLSKHULQJ HDUO\ KXPDQ KLVWRU\ XVLQJ $SSUR[LPDWH %D\HVLDQ &RPSXWDWLRQ DQG  ZKROH JHQRPHV IURP &HQWUDO DQG 6RXWKHUQ $IULFD 0DQXVFULSW ,9 %UHWRQ * %DUKDP / 0XGHQGD * 6RRG\DOO + 6FKOHEXVFK &0

-DNREVVRQ 0  7KH ³%D7ZD´ SRSXODWLRQV IURP UHPRWH DUHDV LQ =DPELD UHWDLQ DQFHVWU\ RI SDVW IRUDJHU JURXSV 0DQXVFULSW HTXDO FRQWULEXWLRQ

(6)
(7)

&RQWHQWV

 5HVHDUFK DLPV                                                                                               

 %DFNJURXQG RQ JHQHWLFV DQG SRSXODWLRQ JHQHWLFV                                     

 7KH EDVHV RI JHQHWLFV JHQRPH RUJDQL]DWLRQ DQG SRSXODWLRQ

JHQHWLFV                                                                                                  +LVWRULFDO YLJQHWWHV                                                                  $Q LQWURGXFWLRQ WR WKH KXPDQ JHQRPH                                 *HQHWLF YDULDWLRQ                                                                     7KH IDWH RI D PXWDWLRQ                                                             3RSXODWLRQ JHQHWLFV LQ D QXWVKHOO                                           

 7HOOLQJ WKLQJV DSDUW LQ SRSXODWLRQ JHQHWLFV                           

 +RZ WR UHDG JHQHWLF LQIRUPDWLRQ                                                      

 6RPH KLVWRU\                                                                           

 613 DUUD\V                                                                             

 +LJKWKURXJKSXW VHTXHQFLQJ PHWKRGV                                 

 $Q LQWURGXFWLRQ WR SURFHVVLQJ +76 GDWD                              

 )XUWKHU SURFHVVLQJ RI JHQRPHZLGH YDULDQW GDWD                

 '1$ IURP H[WDQW DQG DQFLHQW SRSXODWLRQV                          

 'HVFULELQJ JHQHWLF GLYHUVLW\                                                               

 6XPPDU\ VWDWLVWLFV                                                                 

 9LVXDO H[SORUDWLRQ RI JHQHWLF GDWD                                         

 &OXVWHULQJ PHWKRGV                                                                

 ,QIHUULQJ GHPRJUDSKLF KLVWRU\                                                           

 'LYHUJHQFH                                                                              

 0LJUDWLRQ                                                                                

 (IIHFWLYH SRSXODWLRQ VL]H                                                        

 $SSUR[LPDWH %D\HVLDQ &RPSXWDWLRQ ± D YHUVDWLOH

IUDPHZRUN                                                                               

 $ VWXG\ VSHFLHV +RPR VDSLHQV                                                                  

 $Q LQWHUGLVFLSOLQDU\ DSSURDFK                                                            

 'LYHUVLW\ RI SURFHVVHV LPSDFWLQJ WKH JHQRPH                      

 (YLGHQFH IURP RWKHU ILHOGV                                                     

 $EXQGDQFH RI UHVRXUFHV                                                                    

 6RPH DVSHFWV RI WKH KXPDQ UHIHUHQFH JHQRPH                     

 5HIHUHQFH ILOHV DQG GDWDVHWV                                                   

(8)

 (WKLFDO DVSHFWV                                                                                     

 (WKLFDO FRQFHUQV LQ HYROXWLRQDU\ JHQHWLFV                            

 7KH GLIIHUHQW VWDJHV RI D UHVHDUFK SURMHFW  IRFXV RQ

VDPSOLQJ                                                                                  

 ,VVXHV RI QDPLQJ DQG FDWHJRUL]DWLRQ                                    

 :KDW ZH NQRZ DERXW +RPR VDSLHQV KLVWRU\ ZLWK D IRFXV RQ $IULFD        

 +XPDQ RULJLQV LQ $IULFD                                                                     

 3UHIDUPLQJ SRSXODWLRQ VWUXFWXUH LQ 6XE6DKDUDQ $IULFD                 

 *HQH IORZ HYHQWV PRGLILHG WKH SRSXODWLRQ VWUXFWXUH                        

 6XPPDU\ DVVHPEOLQJ D FRPSDUDWLYH GDWDVHW IRU VWXGLHV RI

6XE6DKDUDQ $IULFDQ GHPRJUDSKLF KLVWRU\                                      

 (VWLPDWLQJ GLYHUJHQFH WLPHV DQG PRGHOV RI PRGHUQ KXPDQ

HYROXWLRQ                                                                                              

 6XPPDU\ RI WKH SDSHUV                                                                                

  .KRH6DQ JHQRPHV XQLTXH YDULDWLRQ GHHS GLYHUJHQFH

FKDQJHV LQ SRSXODWLRQ VL]H DQG DGDSWDWLRQ 3DSHU ,                        

 3URFHVVLQJ KLJKFRYHUDJH JHQRPHV WR VWXG\ KXPDQ HYROXWLRQ

3DSHUV ,, DQG ,,,                                                                                

  JHQRPHV IURP &HQWUDO $IULFD DQG HYROXWLRQDU\ KLVWRU\ RI

6XE6DKDUDQ $IULFD 3DSHU ,,,                                                          

 $ QHZ SLHFH WR WKH SX]]OH DXWRVRPDO GLYHUVLW\ IURP WZR %D7ZD =DPELDQ DERULJLQDO SRSXODWLRQV 3DSHU ,9                                      

 &RQFOXVLRQV DQG IXWXUH SURVSHFWV                                                                

 6YHQVN VDPPDQIDWWQLQJ                                                                               

 5pVXPp HQ IUDQoDLV                                                                                      

 $FNQRZOHGJPHQWV                                                                                        

(9)

$EEUHYLDWLRQV

$%& $SSUR[LPDWH %D\HVLDQ &RPSXWDWLRQ

$' $QQR GRPLQL

D'1$ DQFLHQW '1$

$0+ $QDWRPLFDOO\ PRGHUQ KXPDQV

ES %DVH SDLU

%465 %DVH TXDOLW\ VFRUH UHFDOLEUDWLRQ

&$5 &HQWUDO $IULFDQ 5HSXEOLF

&19 &RS\QXPEHU YDULDWLRQ

'5& 'HPRFUDWLF 5HSXEOLF RI WKH &RQJR

*$7. *HQRPH $QDO\VLV 7RRONLW +*'3 +XPDQ *HQRPH 'LYHUVLW\ 3URMHFW +76 +LJKWKURXJKSXW VHTXHQFLQJ +:( +DUG\:HLQEHUJ HTXLOLEULXP .6 .KRH6DQ N\D 7KRXVDQG \HDUV DJR /' /LQNDJH GLVHTXLOLEULXP 0/ 0D[LPXP OLNHOLKRRG PW'1$ 0LWRFKRQGULDO '1$ 1H (IIHFWLYH SRSXODWLRQ VL]H 22$ 2XWRI$IULFD HYHQW 3& 3ULQFLSDO FRPSRQHQW

3&$ 3ULQFLSDO FRPSRQHQW DQDO\VLV

6)6 6LWH IUHTXHQF\ VSHFWUXP

6*'3 6LPRQ¶V *HQRPH 'LYHUVLW\ 3URMHFW

613 6LQJOH QXFOHRWLGH SRO\PRUSKLVP

5+* 5DLQIRUHVW KXQWHUJDWKHUHU

5+*Q 5DLQIRUHVW KXQWHUJDWKHUHU QHLJKERXU

52+ 5XQV RI KRPR]\JRVLW\

705&$ 7LPH WR WKH PRVW UHFHQW FRPPRQ DQFHVWRU

77 7ZR7ZR PHWKRG

9465 9DULDQW TXDOLW\ VFRUH UHFDOLEUDWLRQ

(10)
(11)

 5HVHDUFK DLPV

7KH DLP RI P\ WKHVLV LV WR GHVFULEH KXPDQ JHQHWLF GLYHUVLW\ ZLWK D IRFXV RQ 6XE6DKDUDQ $IULFDQ SRSXODWLRQV DQG WR XVH LW WR GHFLSKHU WKH SRSXODWLRQ KLV WRU\ RI PRGHUQ KXPDQV 0RUH VSHFLILFDOO\ WKH DLPV ZHUH WR

, 3UHSDUH KLJKFRYHUDJH JHQRPH GDWD WR DQVZHU TXHVWLRQV RI UHOHYDQFH

LQ SRSXODWLRQV XQGHUUHSUHVHQWHG LQ JHQHWLF GDWDEDVHV E\ VXUYH\LQJ WKH OLWHUDWXUH DQG GHYHORSLQJ DQG WHVWLQJ D SLSHOLQH 7KLV ZDV GRQH IRU WKH DXWRVRPHV 3DSHU ,, DQG IRU WKH VH[ FKURPRVRPHV DQG PLWRFKRQGULDO '1$ 3DSHU ,,, 

,, 2EWDLQ HVWLPDWHV RI JHQHWLF GLYHUVLW\ PHDVXUHV DQG GHVFULEH UHODWLRQ

VKLSV EHWZHHQ SRSXODWLRQV XVLQJ QRQDVFHUWDLQHG PDUNHUV 3DSHUV , DQG ,,, 

,,, ,QYHVWLJDWH WKH GHHS SRSXODWLRQ KLVWRU\ LQ 6XE6DKDUDQ $IULFD LQ SDU

WLFXODU WKH WRSRORJ\ RI WKH WUHH RI PRGHUQ KXPDQV ZLWK KLJKFRYHUDJH JHQRPHV IURP VRXWKHUQ DQG FHQWUDO $IULFD 3DSHUV , DQG ,,, 

,9 'HVFULEH JHQHWLF GLYHUVLW\ LQ NH\ SRSXODWLRQV WKDW KDYH QRW EHHQ VH

TXHQFHG HDUOLHU DQG H[SORUH WKHLU UHODWLRQVKLSV ZLWK RWKHU SRSXODWLRQV DQG WKHLU KLVWRU\ 3DSHUV ,,, DQG ,9 

9 3HUIRUP DQ $SSUR[LPDWH %D\HVLDQ &RPSXWDWLRQ DQDO\VLV RQ HQWLUH

JHQRPHV WR LQIHU WKH SRSXODWLRQ KLVWRU\ RI VXE6DKDUDQ JURXSV 3DSHU ,,, 

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

(12)

 %DFNJURXQG RQ JHQHWLFV DQG SRSXODWLRQ

JHQHWLFV

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

,Q WKLV WKHVLV , LQYHVWLJDWH YDULRXV DVSHFWV RI KXPDQ GHPRJUDSK\ , XVH KHUH D SRSXODWLRQ JHQHWLFV GHILQLWLRQ RI GHPRJUDSK\ XQGHUVWRRG DV QHXWUDO SURFHVVHV WKDW FDQ DIIHFW WKH JHQHWLF GLYHUVLW\ RI D SRSXODWLRQ VXFK DV FKDQJHV LQ SRSXODWLRQ VL]H GULIW DQG SRSXODWLRQ VWUXFWXUH 7KLV LV GLIIHUHQW EXW UHODWHG WR WKH PRUH VWDQGDUG GHILQLWLRQ RI GHPRJUDSK\ DV WKH VWXG\ RI SRSXODWLRQV¶ OLIH KLVWRU\ WUDLWV YLD VWDWLVWLFV VXFK DV ELUWK DQG GHDWK UDWHV 1DWXUDO VHOHFWLRQ LV D QRQQHXWUDO SURFHVV ZKLFK DOVR DIIHFWV JHQHWLF GLYHUVLW\ LQGLYLGXDOV¶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

+HUH , VWDUW ZLWK DQ LQWURGXFWLRQ WR JHQHWLFV DQG SRSXODWLRQ JHQHWLFV , WKHQ SUHVHQW GLIIHUHQW PHWKRGV WR JHQHUDWH JHQHWLF GDWD DV ZHOO DV WRROV DQG PHWKRGV WR GHVFULEH JHQHWLF GLYHUVLW\ DQG WR LQIHU GHPRJUDSKLF KLVWRU\ VHFRQG WKLUG DQG IRXUWK ER[HV LQ )LJXUH  

(13)

Sample modern human populations Build a dataset of genome-wide variation Identify the demographic model best explaining the data

sequencing individual 1A ACTTCGTCACG individual 1B ACCTCGTCATG individual 2A ACCTCGTCACG individual 2B ACTTCGTCATG Describe genetic diversity agriculturalist pastoralist hunter-gatherer

First axis of variance Second axis

of variance population

population

populations and

present

)LJXUH  +XPDQ SRSXODWLRQ JHQHWLFV VWXGLHV D JUDSKLFDO DEVWUDFW

 7KH EDVHV RI JHQHWLFV JHQRPH RUJDQL]DWLRQ DQG

SRSXODWLRQ JHQHWLFV

 +LVWRULFDO YLJQHWWHV

7ZR VHPLQDO ZRUNV IRU XQGHUVWDQGLQJ HYROXWLRQDU\ ELRORJ\ ZHUH SXEOLVKHG LQ

WKH WKFHQWXU\ &KDUOHV 'DUZLQ¶V ³7KH 2ULJLQ RI 6SHFLHV´ LQ  'DUZLQ

 DQG *UHJRU 0HQGHO¶V ³([SHULPHQWV LQ 3ODQW +\EULGV´ LQ  0HQGHO   7KH ILUVW IRFXVHG RQ VHOHFWLYH IRUFHV VKDSLQJ GLYHUVLW\ ZKLOH WKH VHFRQG HVWDEOLVKHG WKH ODZV RI 0HQGHOLDQ LQKHULWDQFH

'DUZLQ GHVFULEHG PDQ\ H[DPSOHV RI YDULDWLRQ ZLWKLQ DQG EHWZHHQ VSHFLHV WR VXSSRUW WKH IDFW RI HYROXWLRQ DQG SURSRVHG D WKHRU\ WR H[SODLQ LW QDWXUDO VHOHFWLRQ LH WKH LGHD WKDW LQGLYLGXDOV ZLWK DGYDQWDJHRXV IHDWXUHV KDYH PRUH RIIVSULQJ WKDQ RWKHU LQGLYLGXDOV DQG WKXV WKDW WKH SRSXODWLRQ FKDQJHV IURP JHQ HUDWLRQ WR JHQHUDWLRQ JLYHQ VRPH FRQGLWLRQV YDULDWLRQ LQ WKH SRSXODWLRQ DQG OLPLWHG UHVRXUFHV  +LV ZRUN KDV EHHQ UHILQHG EXW PDQ\ RI WKH RULJLQDO LGHDV UHPDLQ LQ PRGHUQ WKHRU\ 7KH DFFHSWHG PHFKDQLVP IRU KHUHGLW\ DW WKDW WLPH ZDV WKDW SDUHQWDO WUDLWV ZHUH PL[HG DQG ³GLOXWHG´ LQ RIIVSULQJ ³EOHQGLQJ KHUHG LW\´  WKLV UHSUHVHQWHG D GLIILFXOW\ IRU 'DUZLQ DV LW GLG QRW ILW ZLWK KLV REVHU 

(14)

YDWLRQV 7KLV ZDV WR EH VROYHG ZLWK 0HQGHO¶V ZRUN WKDW VKRZHG WKDW KHUHGLW\ ZDV EDVHG RQ GLVFUHWH XQLWV WKDW GR QRW EOHQG

,Q PRGHUQ WHUPV WKHVH XQLWV FDQ EH FDOOHG JHQHV RU ORFL VLQJXODU ORFXV  'LIIHUHQW YHUVLRQV RI D JHQH RU D ORFXV DUH FDOOHG DOOHOHV 7KH JHQRW\SH LV WKH DVVRUWPHQW RI DOOHOHV DW D ORFXV ZKLOH WKH SKHQRW\SH LV DQ REVHUYDEOH SK\VLFDO WUDLW ± VXFK DV H\H FRORU 2QH IRFXV RI JHQHWLFV LV WR LGHQWLI\ WKH JHQRW\SHV XQGHUO\LQJ D VSHFLILF SKHQRW\SH

0HQGHO IRUPXODWHG WKUHH ODZV ZKLFK GHVFULEH 0HQGHOLDQ WUDLWV ± WKDW LV SKHQRW\SLF WUDLWV WKDW DUH GHSHQGHQW RQ D VLQJOH JHQH IRU H[DPSOH ODFWDVH

SHUVLVWHQFH WKH DELOLW\ WR GLJHVW ODFWRVH DV DQ DGXOW  7KH ³/DZ RI 6HJUHJD

WLRQ´ VWDWHV WKDW HDFK GLSORLG LQGLYLGXDO KDV WZR JHQHWLF FRSLHV RI HDFK ORFXV DQG WKDW WKHVH FRSLHV VHJUHJDWH GXULQJ PHLRVLV FHOO GLYLVLRQV WKDW UHVXOW LQ JD PHWHV LH VSHUP DQG HJJ FHOOV  *DPHWHV KDYH RQO\ RQH JHQHWLF FRS\ RI HDFK ORFXV DQG WKH RIIVSULQJ UHFHLYHV RQH FRS\ IURP HDFK RI LWV SDUHQWV 7KH ³/DZ RI ,QGHSHQGHQW $VVRUWPHQW´ VWDWHV WKDW DOOHOHV IRU GLIIHUHQW WUDLWV DUH LQKHULWHG LQGHSHQGHQWO\ 7KDW LV RQO\ SDUWLDOO\ WUXH DV VRPH DOOHOHV DUH ³OLQNHG´ WR RWK HUV  7KH ³/DZ RI 'RPLQDQFH´ KDV WR GR ZLWK WKH GRPLQDQFH RI RQH DOOHOH RYHU WKH RWKHU ± IRU H[DPSOH LQ WKH FDVH RI ODFWDVH SHUVLVWHQFH WKH DOOHOH ³GLJHVWHU´ LV GRPLQDQW FRPSDUHG WR WKH ³QRQGLJHVWHU´ DOOHOH DQG DQ LQGLYLGXDO ZLWK D VLQJOH ³GLJHVWHU´ DOOHOH ZLOO H[SUHVV WKH SKHQRW\SH ³GLJHVWHU´ +RZHYHU QRW DOO WUDLWV VKRZ WKLV W\SH RI GRPLQDQFH 0DQ\ WUDLWV DUH QRW 0HQGHOLDQ DQG DUH FDOOHG ³FRPSOH[ WUDLWV´ +HLJKW RU VWDWXUH LV VXFK D WUDLW ,W ZDV VKRZQ WKDW WKRXVDQGV RI YDULDQWV H[SODLQ D OLWWOH ELW RI WKH YDULDWLRQ LQ DGXOW KHLJKW ZKLFK LV DOVR LQIOXHQFHG E\ WKH HQYLURQPHQW VXFK DV GLHW <HQJR HW DO  

 $Q LQWURGXFWLRQ WR WKH KXPDQ JHQRPH

'HR[\ULERQXFOHLF DFLG RU '1$ ZDV VKRZQ WR EH WKH KHULWDEOH PDWHULDO LQ  $YHU\ HW DO  DQG LWV GRXEOHKHOLFDO VWUXFWXUH ZDV GHVFULEHG OHVV WKDQ WHQ \HDUV ODWHU :DWVRQ DQG &ULFN   '1$ FRQVLVWV RI D VHTXHQFH RI PROHFXOHV FDOOHG ³QXFOHRWLGHV´ RIWHQ GHVLJQDWHG DV RQH RI IRXU OHWWHUV $ 7 & RU *  '1$ H[LVWV PRVWO\ DV D SDLUHG GRXEOH VWUDQG KHQFH WKH XQLW EDVH SDLU RU ³ES´ D EDVH SDLU LV HLWKHU ³$7´ ³7$´ ³&*´ RU ³*&´ 

,Q HXNDU\RWHV ± RUJDQLVPV ZKRVH FHOOV FRQWDLQ D QXFOHXV  '1$ LV RUJDQ LVHG LQ FKURPRVRPDO XQLWV ZKLFK DUH PRUH RU OHVV FRQGHQVHG GXULQJ WKH FHOO F\FOH DQG DUH ORFDWHG LQ WKH QXFOHXV +RZ H[DFWO\ WKH JHQHWLF LQIRUPDWLRQ LV RUJDQLVHG YDULHV DFURVV HXNDU\RWLF VSHFLHV ,Q KXPDQV WKHUH DUH  SDLUV RI FKURPRVRPHV LQ DOPRVW DOO FHOOV  RI WKHVH SDLUV DUH WZR YHUVLRQV RI KLJKO\ VLPLODU FKURPRVRPHV +XPDQV DUH GLSORLG WZR FRSLHV RI WKH FKURPRVRPHV

DQG FKURPRVRPHV  WR  DUH WKH DXWRVRPHV 7KH UGSDLU RI FKURPRVRPHV

FRUUHVSRQGV WR WKH VH[ FKURPRVRPHV IHPDOHV KDYH WZR ; FKURPRVRPHV ZKLOH PDOHV KDYH RQH ; DQG RQH < FKURPRVRPH $GGLWLRQDOO\ WKH PLWRFKRQGULD ZKLFK LV LQYROYHG LQ HVVHQWLDO PHWDEROLF DFWLYLWLHV DQG LV SUHVHQW LQ PDQ\ FRSLHV 

(15)

LQ KXPDQ FHOOV DOVR FRQWDLQV '1$ WKH PLWRFKRQGULDO '1$ PW'1$  7R JHWKHU WKH DXWRVRPHV WKH VH[ FKURPRVRPHV DQG WKH YHU\ VKRUW PW'1$ IRUP WKH ³KXPDQ JHQRPH´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

 *HQHWLF YDULDWLRQ

7KH '1$ FRQWDLQV JHQHV PRVW RI ZKLFK FRGH IRU SURWHLQV DV ZHOO DV QRQ FRGLQJ UHJLRQV VRPH RI ZKLFK DUH LPSRUWDQW IRU WKH UHJXODWLRQ RI JHQH H[

SUHVVLRQ ,Q KXPDQV∼ RI WKH JHQRPH LV RFFXSLHG E\ SURWHLQFRGLQJ JHQHV

-REOLQJ HW DO   9DULDWLRQ LQ SURWHLQFRGLQJ JHQHV EXW DOVR LQ RWKHU JH QHWLF UHJLRQV LPSDFWV WKH SKHQRW\SH RI DQ LQGLYLGXDO :KLOH D JHQH JHQHUDOO\ FRUUHVSRQGV WR VHYHUDO KXQGUHGV RI EDVHV D ORFXV FDQ FRUUHVSRQG WR D VLQJOH EDVH

:KHQ DW D JLYHQ SRVLWLRQ D FKURPRVRPH KDV DQ ³$´ ZKLOH DQRWKHU KDV D ³&´ ZH FDOO LW D 613 ³VLQJOH QXFOHRWLGH SRO\PRUSKLVP´ $ JHQHWLF SRO\

PRUSKLVP LV D ORFXV IRU ZKLFK WKHUH DUH DW OHDVW WZR YDULDQWV ,W LV ELDOOHOLF LI

WKHUH DUH WZR DOOHOHV DQG PXOWLDOOHOLF LI WKHUH DUH PRUH WKDQ WZR ,QVHUWLRQV DQG GHOHWLRQV  WRJHWKHU FDOOHG LQGHOV  DUH DQRWKHU W\SH RI YDULDQW LQ ZKLFK DQ DOOHOH KDV RQH RU VHYHUDO EDVH SDLUV LQVHUWHG RU GHOHWHG FRPSDUHG WR DQRWKHU DOOHOH DW WKH VDPH ORFXV 0LFURVDWHOOLWHV DUH VKRUW VHTXHQFHV RI QXFOHRWLGHV ZKLFK DUH UHSHDWHG LQ D URZ WKH H[DFW QXPEHU RI UHSHDWV GLIIHU EHWZHHQ DOOHOHV 6WUXF

WXUDO YDULDWLRQ GHVLJQDWHV ODUJH VFDOH ODUJHU WKDQ  EDVH SDLUV JHQRPH

UHDUUDQJHPHQWV VXFK DV LQYHUVLRQV RU FRS\QXPEHU YDULDWLRQ &19  &19V DUH UHJLRQV RI WKH JHQRPH ZKLFK DUH UHSHDWHG DQG WKH QXPEHU RI UHSHDWV YDULHV EHWZHHQ LQGLYLGXDOV $Q H[DPSOH LV WKH DP\ODVH JHQH $0< LQ KXPDQV WKH DP\ODVH LV DQ HQ]\PH SUHVHQW LQ WKH VDOLYD DQG LQYROYHG LQ WKH GHJUDGDWLRQ RI 

(16)

VWDUFK DQG LW KDV EHHQ VKRZQ WKDW WKH QXPEHU RI FRSLHV RI WKH DP\ODVH JHQH YDULHV EHWZHHQ SRSXODWLRQV DQG WKDW LW LV KLJKHU LQ SRSXODWLRQV ZKR WUDGLWLRQ DOO\ KDG D VWDUFKULFK GLHW 3HUU\ HW DO  

1HZ YDULDQWV DSSHDU E\ SURFHVVHV VXFK DV PXWDWLRQ UHFRPELQDWLRQ RU GX SOLFDWLRQ IRU H[DPSOH GXULQJ '1$ UHSOLFDWLRQ ZKHQ '1$ LV FRSLHG  ,Q WKH

LQILQLWH VLWHV PRGHO D FRPPRQO\ XVHG PRGHO LQ WKH SRSXODWLRQ JHQHWLF ILHOG

D QHZ PXWDWLRQ SURGXFHV D QHZ DOOHOH WKH DVVXPSWLRQ LV WKDW WKHUH DUH HQRXJK PXWDEOH VLWHV WKDW WZR PXWDWLRQV ZLOO QRW RFFXU DW WKH VDPH ORFXV

 7KH IDWH RI D PXWDWLRQ

$ QHZ PXWDWLRQ D FKDQJH LQ WKH '1$ VHTXHQFH LV DW ILUVW SUHVHQW LQ D VLQJOH FRS\ LQ D SRSXODWLRQ ,I WKLV QHZ PXWDWLRQ LV QHXWUDO LH LI LW GRHV QRW LPSDFW WKH ILWQHVV RI LWV FDUULHU LW KDV QR HIIHFW RQ WKH QXPEHU RI IHUWLOH RIIVSULQJ RI WKH FDUULHU  WKH SUREDELOLW\ RI WKH PXWDWLRQ WR EH WUDQVPLWWHG WR WKH QH[W JHQHUDWLRQ LV HTXDO WR WKH SUREDELOLW\ RI LW QRW EHLQJ WUDQVPLWWHG LH ORVW  ,I LW LV QRW QHXWUDO LH LW LPSDFWV WKH FDUULHU¶V ILWQHVV WKH FKDQFHV RI LW EHLQJ WUDQVPLWWHG RU ORVW DUH QRW HTXDO DQ\PRUH ,I WKH PXWDWLRQ LV WUDQVPLWWHG WKH VDPH KDSSHQV DW WKH QH[W JHQHUDWLRQ 'HSHQGLQJ RQ WKH QXPEHU RI RIIVSULQJ FDUU\LQJ WKH PXWDWLRQ WKH QXPEHU RI FRSLHV RI WKH QHZ PXWDWLRQ LQFUHDVHV RU GHFUHDVHV LQ WKH SRSXODWLRQ (YHQWXDOO\ LW FDQ EHFRPH IL[HG DOO LQGLYLGXDOV KDYH WKLV VSHFLILF PXWDWLRQ RU DOOHOH DW WKH ORFXV RU ORVW

1HZ PXWDWLRQV DUH WKH SUHUHTXLVLWH IRU JHQHWLF YDULDWLRQ DQG JHQHWLF YDULD WLRQ LV WKH SUHUHTXLVLWH IRU HYROXWLRQ , ZLOO QRZ SUHVHQW KRZ SRSXODWLRQ JHQHW LFV FDQ KHOS XV WR XQGHUVWDQG JHQHWLF YDULDWLRQ DQG HYROXWLRQ

 3RSXODWLRQ JHQHWLFV LQ D QXWVKHOO

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

7KH UHODWLRQVKLS EHWZHHQ JHQRW\SH IUHTXHQFLHV DQG DOOHOH IUHTXHQFLHV LV D FHQWUDO FRQFHSW LQ SRSXODWLRQ JHQHWLFV $W HTXLOLEULXP WKH +DUG\:HLQEHUJ SULQFLSOH GHVFULEHV WKLV UHODWLRQVKLS ,I ZH FRQVLGHU D ELDOOHOLF ORFXV ZLWK DOOHOHV ³$´ DQG ³D´ WKHUH DUH WKUHH SRVVLEO\ JHQRW\SHV DQ LQGLYLGXDO FDQ KDYH WZR ³$´ DOOHOHV JHQRW\SH $$  WZR ³D´ DOOHOHV DD  RU RQH RI HDFK $D  *HQRW\SHV $$ DQG DD DUH KRPR]\JRXV ZKLOH JHQRW\SH $D LV KHWHUR]\JRXV :H GHILQH

p DV WKH IUHTXHQF\ RI DOOHOH ³$´ LQ WKH SRSXODWLRQ EHFDXVH WKHUH DUH RQO\ WZR

(17)

DOOHOHV WKH IUHTXHQF\ RI ³D´ LV q = 1 − p $W +DUG\:HLQEHUJ HTXLOLEULXP

+:(  WKH H[SHFWHG IUHTXHQF\ RI WKH JHQRW\SH $$ LVp×p = p2 WKH H[SHFWHG

IUHTXHQF\ RI WKH JHQRW\SH DD LVq2 = (1 − p)2 DQG WKH H[SHFWHG IUHTXHQF\ RI

WKH JHQRW\SH $D LV2pq = 2p(1 − p) $ FRPPRQ SURFHGXUH LV WR FRPSDUH WKH

REVHUYHG DQG WKH H[SHFWHG JHQRW\SH IUHTXHQFLHV DW D ORFXV LI WKH\ GLIIHU LW PHDQV WKDW WKH SRSXODWLRQ LV QRW DW HTXLOLEULXP DW WKLV ORFXV ZKLFK FDQ EH GXH WR GLIIHUHQW UHDVRQV VXFK DV GHPRJUDSKLF RU DGDSWLYH SURFHVVHV

7KH ILHOG RI SRSXODWLRQ JHQHWLFV VWDUWHG LQ WKH V DV D WKHRUHWLFDO ILHOG VLQFH QR DSSURSULDWH GDWD ZDV DYDLODEOH DW WKH WLPH WR WHVW WKH IRUPXODWHG K\ SRWKHVHV &RQVHTXHQWO\ SRSXODWLRQ JHQHWLF PRGHOV DUH RIWHQ GHYHORSHG IRU ³LGHDO´ LQ D PDWKHPDWLFDO VHQVH SRSXODWLRQV 7KH :ULJKW)LVKHU PRGHO :)

PRGHO )LVKHU  :ULJKW   LV DQ H[DPSOH RI VXFK D PRGHO ,W LV

ZLGHO\ XVHG WR GHVFULEH KRZ DOOHOH IUHTXHQFLHV FKDQJH IURP JHQHUDWLRQ WR JHQ HUDWLRQ LQ SDUWLFXODU LW GHVFULEHV JHQHWLF GULIW , PHQWLRQHG HDUOLHU WKDW 'DU ZLQ¶V WKHRU\ RI HYROXWLRQ KDG EHHQ UHILQHG LQ IDFW LW EHFDPH FOHDU WKDW QDWXUDO VHOHFWLRQ LV QRW WKH RQO\ IRUFH VKDSLQJ JHQHWLF GLYHUVLW\ WKHUH DUH RWKHU IRUFHV DV ZHOO VXFK DV PXWDWLRQ UHFRPELQDWLRQ PLJUDWLRQ DQG JHQHWLF GULIW *HQHWLF GULIW LV WKH ³UDQGRP FKDQJH RI DOOHOH IUHTXHQFLHV LQ SRSXODWLRQV RI ILQLWH VL]H´ 1LHOVHQ DQG 6ODWNLQ   7KH UHODWLYH LPSRUWDQFH RI VHOHFWLRQ DQG GULIW LQ VKDSLQJ JHQHWLF GLYHUVLW\ ZLWKLQ DQG EHWZHHQ VSHFLHV KDV EHHQ WKH VXEMHFW RI PXFK GHEDWH $FFRUGLQJ WR WKH QHXWUDO WKHRU\ RI PROHFXODU HYROXWLRQ .LPXUD   VHOHFWLRQ DORQH LV QRW VXIILFLHQW WR H[SODLQ WKH SDWWHUQV RI JHQHWLF YDULD WLRQ LQ IDFW WKH PDMRULW\ RI PXWDWLRQV DUH QHXWUDO RU VOLJKWO\ GHOHWHULRXV

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

(18)

$Q LPSRUWDQW ZD\ LQ ZKLFK WKH :) PRGHO LV XVHG LV WR GHILQH WKH HIIHFWLYH

SRSXODWLRQ VL]H 1H  *LYHQ D QDWXUDO SRSXODWLRQ ZLWK D FHUWDLQ FHQVXV VL]H QXPEHU RI LQGLYLGXDOV  WKH HIIHFWLYH SRSXODWLRQ VL]H LV WKH FHQVXV VL]H RI D :) SRSXODWLRQ ZLWK WKH VDPH DPRXQW RI JHQHWLF YDULDWLRQ JLYHQ WKH VDPH PXWDWLRQ UDWH )RU :ULJKW)LVKHU SRSXODWLRQV FHQVXV DQG HIIHFWLYH SRSXOD WLRQ VL]H DUH LGHQWLFDO ,Q JHQHUDO WKH FHQVXV VL]H LV ODUJHU WKDQ WKH HIIHFWLYH SRSXODWLRQ VL]H DV SURFHVVHV VXFK DV QRQUDQGRP PDWLQJ UHGXFH WKH QXPEHU RI LQGLYLGXDOV ZKR HIIHFWLYHO\ FRQWULEXWH WR WKH QH[W JHQHUDWLRQ

7KH FRDOHVFHQW .LQJPDQ  LV DQRWKHU NH\ FRQFHSW LQ SRSXODWLRQ JH QHWLFV ,W TXLFNO\ EHFDPH FHQWUDO EHFDXVH LW HIILFLHQWO\ UHODWHV WKHRU\ DQG GDWD ZKLFK LV PRUH DQG PRUH DEXQGDQW VHH 6HFWLRQ  DERXW REWDLQLQJ JHQHWLF GDWD  7KH FRQFHSW RI WKH FRDOHVFHQW LV SRZHUIXO EHFDXVH LW IXQFWLRQV EDFN ZDUG LQ WLPH JLYHQ WZR OLQHDJHV REVHUYHG DW SUHVHQW LW JLYHV WKH SUREDELOLW\ WKDW WKHVH WZR OLQHDJHV FRDOHVFH ± LH KDYH D FRPPRQ DQFHVWRU ±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

 7HOOLQJ WKLQJV DSDUW LQ SRSXODWLRQ JHQHWLFV

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

(19)

RI VHOHFWLRQ DUH ORFDO DURXQG WKH ORFXV XQGHU VHOHFWLRQ ZKLOH GHPRJUDSK\ LP SDFWV WKH JHQRPH PRUH JHQHUDOO\ ,Q WKH SDVW PXFK HIIRUW KDV EHHQ SXW LQWR VHOHFWLQJ QHXWUDO UHJLRQV IRU GHPRJUDSKLF LQIHUHQFHV 5RVHQEHUJ HW DO  9HUGX HW DO   7RGD\ ZH KDYH HDVLHU DFFHVV WR JHQRPHZLGH GDWD DQG WKXV WKH LVVXH PLJKW VHHP OHVV UHOHYDQW +RZHYHU PDQ\ UHFHQW VWXGLHV KDYH VKRZQ WKDW WKH VLWXDWLRQ LV QRW VROYHG ZLWK L LQFUHDVLQJ HYLGHQFH WKDW PRVW RI WKH VHOHFWLRQ SURFHVVHV GR QRW RFFXU DV ³VWURQJ VZHHSV´ ZKHUH D VLQJOH QHZ DGYDQWDJHRXV DOOHOH ULVHV WR KLJK IUHTXHQF\ EXW UDWKHU IURP VWDQGLQJ YDULDWLRQ ZKHUH RQH RU VHYHUDO DOOHOHV ULVH LQ IUHTXHQF\ IURP D SRRO RI H[LVWLQJ DOOHOHV DQG LL HYLGHQFH RI ³ORQJUDQJH´ HIIHFWV RI VHOHFWLRQ LH LW LV GLIILFXOW WR ILQG UHJLRQV RI WKH JHQRPH ZKLFK DUH WUXO\ QHXWUDO +HUQDQGH] HW DO  7RUUHV HW DO  

$QRWKHU GLIILFXOW\ LV WR XQWDQJOH SDUDPHWHUV 2QH H[DPSOH LV JLYHQ E\ WKH SDUDPHWHUV ³GLYHUJHQFH WLPH´ DQG ³JHQH IORZ´ :H FDQ FRQVLGHU WKH VLPSOH

FDVH RI WZR SRSXODWLRQV WKDW KDYH GLYHUJHG VRPH WLPH LQ WKH SDVW DWTdiv DQG

WKDW KDYH D JLYHQ DPRXQW RI JHQHWLF GLIIHUHQWLDWLRQ WRGD\ ,I WKH WZR SRSXOD WLRQV GLG QRW H[FKDQJH DQ\ PLJUDQWV VLQFH WKH\ GLYHUJHG DOO WKH JHQHUDWLRQV VLQFH GLYHUJHQFH KDYH FRQWULEXWHG WR WKHLU GLIIHUHQWLDWLRQ %XW LI WKH\ KDYH H[FKDQJHG PLJUDQWV WKLV KDV PDGH WKH WZR SRSXODWLRQV PRUH VLPLODU DJDLQ

&RQVHTXHQWO\ IRU D JLYHQ DPRXQW RI JHQHWLF GLIIHUHQWLDWLRQTdivZLOO EH PRVW

UHFHQW LQ WKH DEVHQFH RI JHQH IORZ DQG ZLOO LQFUHDVH ZLWK WKH DPRXQW RI JHQH IORZ ,I ZH GR QRW FRQVLGHU JHQH IORZ DQG LW KDV LQ IDFW KDSSHQHG  ZH ZLOO XQGHUHVWLPDWH WKH GLYHUJHQFH WLPH

 +RZ WR UHDG JHQHWLF LQIRUPDWLRQ

7KLV LV WKH VWHS LQ D SRSXODWLRQ JHQHWLF VWXG\ ZKHUH D GDWDVHW RI JHQRPHZLGH YDULDWLRQ LV EXLOW IURP '1$ VDPSOHV )LJXUH   7KLV FDQ EH GRQH LQ VHYHUDO ZD\V

 6RPH KLVWRU\

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

(20)

ILUVW JHQRPH WR EH VHTXHQFHG LQ  ZDV WKDW RI D YLUXV WKH ij; EDFWH ULRSKDJH $ FRPELQDWLRQ RI GLIIHUHQW WHFKQLTXHV FXOPLQDWHG LQ WKH SXEOLFDWLRQ RI WZR GUDIW KXPDQ JHQRPHV LQ  WKH SXEOLFO\ IRXQGHG ³PRVDLF´ UHIHUHQFH JHQRPH /DQGHU HW DO   DQG D SULYDWH LQLWLDWLYH OHG E\ - &UDLJ 9HQWHU 9HQWHU HW DO   6LQFH WKHQ ZKROH JHQRPHV KDYH EHHQ VHTXHQFHG IURP D ODUJH QXPEHU RI RUJDQLVPV

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

 613 DUUD\V

)ROORZLQJ WKH +DS0DS 3URMHFW *LEEV HW DO   ZKROHJHQRPH 613 DU

UD\V RU FKLSV ZHUH GHYHORSHG E\ YDULRXV FRPSDQLHV %DVHG RQ WKH KXPDQ

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|P HW DO   7KLUG 613 DUUD\V DUH PRVWO\ PHDQW DV WKHLU QDPH VXJJHVW WR VWXG\ 613V EXW DV ZH KDYH VHHQ HDUOLHU WKHUH DUH PDQ\ RWKHU W\SHV RI JH 

(21)

QHWLF YDULDWLRQ 7KH ILUVW OLPLWDWLRQ FDQ EH SDUWLDOO\ RYHUFRPH ZLWK D WHFKQLTXH FDOOHG ³LPSXWDWLRQ´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

 +LJKWKURXJKSXW VHTXHQFLQJ PHWKRGV

)URP  RQZDUG VHYHUDO QHZ VHTXHQFLQJ WHFKQRORJLHV KHUH UHIHUUHG WR DV

KLJKWKURXJKSXW VHTXHQFLQJ PHWKRGV +76 EXW RIWHQ FDOOHG ³QH[WJHQHUDWLRQ

VHTXHQFLQJ´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³KLJKFRYHUDJH´ JHQRPH FRYHUDJHV DURXQG  ; RU OHVV DUH ³ORZ FRYHUDJH´ :KLOH WKH ODWHVW ,OOXPLQD PDFKLQH WKH 1RYD6HT 6HULHV FDQ VH TXHQFH D KXPDQ JHQRPH LQ DQ KRXU WKH UDZ GDWD QHHGV SURFHVVLQJ EHIRUH LW FDQ EH XVHG IRU DQDO\VHV

(22)

2QJRLQJ WHFKQRORJLFDO FKDOOHQJHV IRFXV RQ VHTXHQFLQJ IURP VLQJOH '1$ PROHFXOHV RU RQ VHTXHQFLQJ ORQJHU IUDJPHQWV 5HXWHU HW DO   ZKLFK ZRXOG DOORZ IRU EHWWHU FDOOLQJ RI LQVHUWLRQV GHOHWLRQV DQG VWUXFWXUDO YDULDQWV DQG DO OHYLDWH WKH LVVXH RI SKDVLQJ GLVFXVVHG EHORZ  0RUHRYHU ZKLOH WKH FRVW RI VHTXHQFLQJ D KXPDQ JHQRPH KDV GHFUHDVHG WUHPHQGRXVO\ DVVRFLDWHG FRVWV VXFK DV GDWD SURFHVVLQJ DQG VWRUDJH DUH RIWHQ RYHUORRNHG /LJKWERG\ HW DO   $QRWKHU SUDFWLFDO LVVXH LV WKH VWDQGDUGL]DWLRQ RI SURFHVVLQJ SLSHOLQHV DQG PHUJLQJ GDWD JHQHUDWHG ZLWK GLIIHUHQW VHTXHQFLQJ SODWIRUPV DQG SURFHVV LQJ SLSHOLQHV %DLFKRR HW DO  5HJLHU HW DO  

 $Q LQWURGXFWLRQ WR SURFHVVLQJ +76 GDWD

$ )$674 ILOH  WKH FRPPRQ IRUPDW IRU UDZ VHTXHQFLQJ GDWD ±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

7KH SURFHVVLQJ SLSHOLQH JHQHUDOO\ EHJLQV ZLWK DOLJQLQJ WKH UHDGV VWRUHG LQ WKH )$674 WR D UHIHUHQFH JHQRPH IRU H[DPSOH ZLWK WKH VRIWZDUH EZD /L DQG 'XUELQ   7KLV LV WKH ³PDSSLQJ VWHS´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¶ VHTXHQFHV WRJHWKHU ZLWK DPRQJ RWKHU LQIRUPDWLRQ WKH FRRUGLQDWHV ZKHUH WKH\ PDS LQ WKH UHIHU HQFH JHQRPH D PDSSLQJ TXDOLW\ IRU HDFK UHDG DQG D VWULQJ GHVFULELQJ KRZ HDFK SRVLWLRQ RI HDFK UHDG DOLJQV WR WKH UHIHUHQFH

%HIRUH YDULDQWV FDQ EH LGHQWLILHG ³YDULDQW FDOOLQJ´  VHYHUDO VWHSV KDYH WR EH SHUIRUPHG RQ %$0 ILOHV 7KH *HQRPH $QDO\VLV 7RRONLW *$7. 'H3ULVWR HW DO  LV D FROOHFWLRQ RI WRROV DQG JXLGHOLQHV GHYHORSHG DW WKH 'DWD 6FL HQFHV 3ODWIRUP DW WKH %URDG ,QVWLWXWH ZKLFK HQDEOH WKH SURFHVVLQJ RI %$0 ILOHV DQG FDOOLQJ RI YDULDQWV ,Q SDUWLFXODU WKH *$7. %HVW 3UDFWLFHV 'H3ULVWR HW DO  9DQ GHU $XZHUD HW DO  DUH VWDQGDUG ZRUNIORZV LQ WKH DQDO\VHV RI +76 GDWD 3DSHU ,, IRFXVHV RQ DSSO\LQJ WKH ³*HUPOLQH VKRUW YDULDQW GLV 

(23)

FRYHU\ 613V  ,QGHOV ´ ZRUNIORZ WR SRSXODWLRQV XQGHUUHSUHVHQWHG LQ JHQHWLF GDWDVHWV ZKLOH GDWD SUHSDUHG IROORZLQJ GLIIHUHQW YHUVLRQV RI WKLV ZRUNIORZ LV DQDO\VHG LQ 3DSHUV , DQG ,,, 6RPH VWHSV FRPPRQO\ LQFOXGHG DUH PDUNLQJ RI GX SOLFDWH UHDGV LH UHDGV ZKLFK FRUUHVSRQG WR WKH VHTXHQFLQJ RI WKH VDPH '1$ IUDJPHQW DQG WKXV FDUU\ UHGXQGDQW LQIRUPDWLRQ  UHDOLJQPHQW DURXQG LQGHOV WKH PDSSLQJ RI UHDGV FORVH WR FRPPRQ LQGHOV LV LPSURYHG  DQG D UHFDOLEUD WLRQ RI WKH EDVH TXDOLW\ VFRUH ³%465´ 

9DULDQW FDOOLQJ LV SHUIRUPHG RQ IXOO\ SURFHVVHG %$0 ILOHV ,Q WKLV WKHVLV YDULDQW FDOOLQJ ZDV SHUIRUPHG ZLWK WZR GLIIHUHQW *$7. YDULDQW FDOOLQJ WRROV ³8QLILHG*HQRW\SHU´ IRU 3DSHU , DQG ³+DSORW\SH&DOOHU´ IRU 3DSHUV ,, DQG ,,, $Q DGYDQWDJH RI +DSORW\SH&DOOHU LV WKDW LW FDOOV 613V DQG LQGHOV VLPXOWDQH RXVO\ DQG SHUIRUPV ORFDO UHDOLJQPHQW LQ UHJLRQV RI WKH JHQRPH WKDW VHHP WR FRQWDLQ YDULDQWV ,W LV DOVR EXLOW WR IDFLOLWDWH WKH DGGLWLRQ RI QHZ VDPSOHV 2WKHU YDULDQW FDOOHUV H[LVW VXFK DV )UHH%D\HV *DUULVRQ DQG 0DUWK  DQG 6$0 WRROV PSLOHXS /L HW DO   )RU D UHYLHZ RI DOLJQPHQW DQG YDULDQW FDOOLQJ DOJRULWKPV VHH 0LHOF]DUHN DQG 6]\GD  

7KH RXWSXW IRUPDW RI YDULDQW FDOOHUV LV D 9&) ILOH 9DULDQW &DOO )RUPDW 'DQHFHN HW DO   9&) ILOHV OLNH 6$0 ILOHV FRQVLVWV RI D KHDGHU DQG D GDWD VHFWLRQ 7KH GDWD VHFWLRQ FDQ FRQWDLQ UHFRUGV IRU L HDFK SRVLWLRQ LQ WKH JHQRPH DQ ³DOO VLWHV 9&)´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¶V VSHFLILF DSSURDFK 9DULDQW 4XDOLW\ 6FRUH 5HFDOLEUDWLRQ  ³9465´  JHQRW\SLQJ PLVVLQJQHVV PLQRU DOOHOH IUHTXHQF\ GHSDUWXUH IURP +:( (QWLUH UHJLRQV ZKLFK DUH NQRZQ WR EH GLIILFXOW WR VHTXHQFH FDQ DOVR EH ILOWHUHG RXW 7KH FKRLFH RI ILOWHULQJ VWHSV GHSHQGV RQ WKH SXUSRVH RI WKH SURMHFW IRU H[ DPSOH PHGLFDO VWXGLHV DUH RIWHQ FRQFHUQHG DERXW IDOVH SRVLWLYHV DQG ILOWHU WKH FDOOVHWV KHDYLO\ ZKLOH VWXGLHV DLPLQJ DW GHVFULELQJ KXPDQ GLYHUVLW\ DYRLG D WRR VWULQJHQW ILOWHULQJ VLQFH LW ZRXOG W\SLFDOO\ ELDV GLYHUVLW\ HVWLPDWHV GRZQZDUGV

 )XUWKHU SURFHVVLQJ RI JHQRPHZLGH YDULDQW GDWD

, KDYH IRFXVHG RQ SUH SURFHVVLQJ RI +76 GDWD DV LW LV D PRUH FRPSOH[ SURFHVV WKDQ SURFHVVLQJ 613 DUUD\ GDWD 7KLV VHFWLRQ SUHVHQWV VRPH SURFHVVLQJ VWHSV ZKLFK FDQ EH DSSOLHG WR YDULDQW GDWD ZKHWKHU LW FRPHV IURP D +76 RU D 613 DUUD\ VWXG\ 1RWH WKDW SUHSURFHVVLQJ VWHSV DUH QHFHVVDU\ IRU 613 DUUD\ GDWD DV ZHOO 3DSHU ,9 JLYHV D SLSHOLQH H[DPSOH 7KH FDOOVHW ILOWHULQJ VWHSV GHVFULEHG LQ WKH SUHFHGLQJ VHFWLRQ UHVHPEOH WKH ILOWHULQJ VWHSV DSSOLHG WR 613 DUUD\ GDWD 

(24)

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

.QRZLQJ WKH SKDVH RI D YDULDQW ± LH ZKLFK KDSORW\SH LW RFFXUV RQ ± LV LPSRUWDQW IRU D QXPEHU RI DQDO\VHV 8QIRUWXQDWHO\ WKH PRVW FRPPRQO\ XVHG +76 WHFKQRORJLHV GR QRW SURYLGH WKDW LQIRUPDWLRQ UHDGV DUH WRR VKRUW WR FRQ WDLQ VHYHUDO YDULDQWV H[FHSW IRU UHJLRQV ZLWK KLJK YDULDQW GHQVLW\  7KLV LQIRU PDWLRQ LV QRW DYDLODEOH IURP 613 DUUD\ GDWD HLWKHU $V D UHVXOW YDULDQW GDWD KDV WR EH SKDVHG 7KHUH DUH WZR PDLQ FDWHJRULHV RI SKDVLQJ DSSURDFKHV 7KH ILUVW FDWHJRU\ LV H[SHULPHQWDO SKDVLQJ UHYLHZHG LQ +XDQJ HW DO    )RU H[DP SOH VRPH +76 WHFKQRORJLHV VXFK DV 3DF%LR 5KRDGV DQG $X   RXWSXW ORQJ UHDGV HQFRPSDVVLQJ VHYHUDO YDULDQWV KDSORW\SHV FDQ EH FRQVWUXFWHG E\ RYHUODSSLQJ VXFK UHDGV 7KH GUDZEDFNV RI WKHVH DSSURDFKHV DUH WKHLU FRVW DQG WKHLU KLJKHU VHTXHQFLQJ HUURU UDWH FRPSDUHG WR VKRUW UHDG VHTXHQFLQJ  +RZ HYHU LW LV OLNHO\ WKDW DV WHFKQRORJLHV LPSURYH WKH\ ZLOO EHFRPH PRUH DQG PRUH SRSXODU 6SHUPW\SLQJ LV DQRWKHU H[SHULPHQWDO DOWHUQDWLYH WKDW SURGXFHV SKDVH GLUHFWO\ DV WKH JHUP FHOOV DUH KDSORLG 7KH VHFRQG FDWHJRU\ LV VWDWLVWLFDO SKDVLQJ  VRPH VRIWZDUHV DUH UHYLHZHG LQ 0LDU HW DO   ,Q WKLV FDVH SKDVLQJ RF FXUV DIWHU WKH VHTXHQFLQJ ,W LV EDVHG RQ WKH LGHD WKDW ZLWKLQ D SRSXODWLRQ WKHUH DUH FOXVWHUV RI OLQNHG YDULDQWV:KHQ FKRRVLQJ D SKDVLQJ DSSURDFK VHYHUDO DV SHFWV KDYH WR EH FRQVLGHUHG VXFK DV FRPSXWLQJ WLPH DFFXUDF\ RI SKDVLQJ DQG DYDLODELOLW\ RI DSSURSULDWH UHIHUHQFH GDWDVHWV LI D UHIHUHQFH GDWDVHW LV QHHGHG  7KH ODWWHU DVSHFW LV SDUWLFXODUO\ UHOHYDQW IRU +76 GDWD DV YDULDQWV WKDW DUH QRW SUHVHQW LQ WKH UHIHUHQFH GDWDVHW ZLOO EH PRUH GLIILFXOW WR SKDVH VHH %HUJVWU|P HW DO  IRU DQ DSSURDFK WR FLUFXPYHQW WKDW LVVXH  ,Q JHQHUDO WKH TXDOLW\ RI WKH LQIHUHQFHV LQFUHDVHV ZLWK WKH VDPSOH VL]H 3RUFX HW DO  

6WDWLVWLFDO SKDVLQJ LV RIWHQ SHUIRUPHG WRJHWKHU ZLWK LPSXWDWLRQ ±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

(25)

RYHUODSSLQJ SRVLWLRQV DUH LPSXWHG LW LV SRVVLEOH WR SHUIRUP WKH DQDO\VHV RQ D ODUJHU QXPEHU RI PDUNHUV

 '1$ IURP H[WDQW DQG DQFLHQW SRSXODWLRQV

%HFDXVH JHQHWLF PDWHULDO LV WUDQVPLWWHG IURP SDUHQW WR RIIVSULQJ WKH JHQRPH RI D JLYHQ LQGLYLGXDO LV LQIRUPDWLYH DERXW LWV DQFHVWRUV 7KLV LV ZK\ JHQRPHV IURP H[WDQW LQGLYLGXDOV FDQ EH XVHG WR LQYHVWLJDWH SDVW SRSXODWLRQ KLVWRU\ $G GLWLRQDOO\ WKH ODVW IHZ GHFDGHV KDYH VHHQ WKH GHYHORSPHQW RI DQFLHQW '1$ D'1$ VHTXHQFLQJ WHFKQLTXHV WR LQYHVWLJDWH SDVW SRSXODWLRQV KLVWRU\ D'1$ GHVLJQDWHV '1$ LVRODWHG IURP ³DQFLHQW´ VSHFLPHQV ZKR GLHG LQ D PRUH RU OHVV UHFHQW SDVW ,W SURYLGHV D ZLQGRZ LQWR VSHFLILF WLPHV UDQJLQJ IURP D FRX SOH RI JHQHUDWLRQV WR KXQGUHG RI WKRXVDQGV RI \HDUV DJR  ,W DOORZV IRU H[ DPSOH WR GHVFULEH WKH JHQHWLF GLYHUVLW\ SULRU WR D SRSXODWLRQ PLJUDWLRQ HYHQW ZKLFK PLJKW KDYH SDUWLDOO\ RU WRWDOO\ UHSODFHG ORFDO SRSXODWLRQV D'1$ LV DOVR YHU\ YDOXDEOH LQ OHDUQLQJ DERXW VSHFLHV WKDW DUH QRW H[WDQW WRGD\ $ IDVFLQDW LQJ H[DPSOH LV WKH D'1$ REWDLQHG IURP DQ H[WLQFW KXPDQ VXEVSHFLHV QDPHG ³'HQLVRYDQ´ DIWHU WKH FDYH ZKHUH LWV UHPDLQV ZHUH ILUVW IRXQG 5HLFK HW DO   ,Q WKDW FDVH D'1$ ZDV H[WUHPHO\ YDOXDEOH DV WKH IHZ UHPDLQV D SKD ODQ[ VRPH WHHWK ZHUH QRW VXIILFLHQW WR GLVWLQJXLVK LW IURP RWKHU KXPDQ UH PDLQV ,W ZDV ODWHU VKRZQ WKDW PRGHUQ KXPDQV LQ SDUWLFXODU LQ 2FHDQLD DQG ,VODQG 6RXWK (DVW $VLD KDYH VXEVWDQWLDO D IHZ SHUFHQW 'HQLVRYDQOLNH DQFHV WU\

$ QXPEHU RI FKDOOHQJHV FKDUDFWHULVH D'1$ VWXGLHV )LUVW LW LV GLIILFXOW WR ILQG JRRG VDPSOHV ERWK HQYLURQPHQWDO FRQGLWLRQV IDYRULQJ WKH SUHVHUYDWLRQ RI '1$ DQG H[FDYDWLRQ HIIRUWV DUH QHFHVVDU\ 6HFRQG LW LV GLIILFXOW DQG FRVWO\ WR VHTXHQFH L WKH VDPSOHV DUH RIWHQ FRQWDPLQDWHG ZLWK H[WUDJHQHRXV '1$

LL WKH '1$ LV GHJUDGHG LQWR VKRUW IUDJPHQWV DQG LLL SK\VLFDO FKHPLFDO DQG

ELRORJLFDO SURFHVVHV FDQ KDYH PRGLILHG WKH VHTXHQFH 7KLUG LW LV QRW HQWLUHO\ FOHDU KRZ WKH W\SLFDO IHDWXUHV RI D'1$ VXFK DV VKRUW IUDJPHQWV ORZ FRYHU DJH VHTXHQFLQJ HUURUV DQG GLIILFXOW\ RI REWDLQLQJ GLSORLG VHTXHQFHV LPSDFW WKH SRSXODWLRQ JHQHWLF IUDPHZRUNV WKDW DUH FRPPRQO\ XVHG DQG KDYH EHHQ GHYHO RSHG IRU PRGHUQ '1$ 0RUHRYHU LW LV UDUH WKDW PRUH WKDQ D IHZ VDPSOHV DUH REWDLQHG IRU D JLYHQ VLWH DW D JLYHQ WLPH 7KLV PDNHV LW GLIILFXOW WR DSSO\ SRSXOD WLRQ EDVHG DQDO\VHV DQG WR NQRZ ZKHWKHU WKH VDPSOHV DUH UHSUHVHQWDWLYH RI WKH SRSXODWLRQ EXW WKLV LV FKDQJLQJ UDSLGO\ ZLWK LQFUHDVLQJ QXPEHU RI JHQRPHV VHTXHQFHG IRU D JLYHQ WLPH SHULRG DQGRU ORFDWLRQ DOORZLQJ IRU ORQJLWXGLQDO VWXGLHV HJ $QWRQLR HW DO  

)RU WKHVH GLIIHUHQW UHDVRQV VWXGLHV EDVHG RQ PRGHUQ '1$ UHPDLQ DQG SUREDEO\ ZLOO UHPDLQ HVVHQWLDO WR RXU XQGHUVWDQGLQJ RI WKH SDVW 7KLV LV SDU WLFXODUO\ WUXH ZLWK VWXGLHV RI KXPDQ KLVWRU\ LQ 6XE6DKDUDQ $IULFD DV KXPDQ UHPDLQV DUH UHODWLYHO\ UDUH LQ 6XE6DKDUDQ $IULFD DQG VXFFHVVIXO D'1$ VWXGLHV HYHQ UDUHU 9LFHQWH DQG 6FKOHEXVFK  

(26)

 'HVFULELQJ JHQHWLF GLYHUVLW\

2QFH LQ SRVVHVVLRQ RI D ZHOOILOWHUHG VHW RI YDULDQWV SRVVLEO\ SKDVHG DQG LP SXWHG UHVHDUFKHUV ZDQW WR H[WUDFW LQIRUPDWLRQ IURP LW WKLUG URZ LQ )LJXUH   $V D ILUVW VWHS ZH ZDQW WR ILQG ZD\V WR ³VXPPDULVH´ WKH GDWD WR LGHQWLI\ LWV PRVW VDOLHQW IHDWXUHV ,Q WKLV VHFWLRQ , SUHVHQW VHYHUDO ZD\V WR GR WKDW

 6XPPDU\ VWDWLVWLFV

6XPPDU\ VWDWLVWLFV FDQ EH XVHG WR FDSWXUH GLIIHUHQW DVSHFWV RI JHQHWLF GLYHU VLW\ 7KH\ FDQ EH FRPSXWHG DW PDQ\ GLIIHUHQW OHYHOV IRU DQ LQGLYLGXDO RU D SRSXODWLRQ VDPSOH IRU WKH HQWLUH JHQRPH IRU VSHFLILF UHJLRQV RU E\ VFDQQLQJ WKH JHQRPH XVLQJ VOLGLQJ ZLQGRZV RU IRU VSHFLILF IXQFWLRQDO FDWHJRULHV IRU H[DPSOH H[RQV LQWURQV RU LQWHUJHQLF UHJLRQV DV LQ 3DSHU ,  , ZLOO SUHVHQW H[ DPSOHV RI VWDWLVWLFV WKDW DUH FDOFXODWHG RQ LQGHSHQGHQW 613V DQG WKDW , XVHG LQ P\ WKHVLV

+HWHUR]\JRVLW\ PHDVXUHV WKH SUREDELOLW\ IRU D VDPSOH RI WZR JHQH FRSLHV WR

KDYH WZR GLIIHUHQW YDULDQWV DW D VLWH 7KLV SUREDELOLW\ LV KLJKHU LQ SRSXODWLRQV ZLWK KLJK JHQHWLF GLYHUVLW\ LI WKH SRSXODWLRQV DUH LQ +:(  2QH ZD\ WR HVWL PDWH KHWHUR]\JRVLW\ LV WR GLYLGH WKH QXPEHU RI KHWHUR]\JRXV VLWHV E\ WKH WRWDO QXPEHU RI VLWHV 7KLV LV FDOOHG REVHUYHG KHWHUR]\JRVLW\ VHH H[DPSOHV LQ 3DSHU ,  $QRWKHU PHDVXUH RI KHWHUR]\JRVLW\ LV H[SHFWHG KHWHUR]\JRVLW\ ,W LV EDVHG RQ WKH +DUG\:HLQEHUJ HTXLOLEULXP 6HFWLRQ   DFFRUGLQJ WR ZKLFK WKH SUR

SRUWLRQ RI KHWHUR]\JRXV LQGLYLGXDOV LV2p(1 − p) ZKHUH p LV WKH IUHTXHQF\ RI

DQ DOOHOH DW D GLSORLG ORFXV 7KLV TXDQWLW\ FDQ EH FDOFXODWHG ORFXV E\ ORFXV DQG DYHUDJHG ,W LV SRVVLEOH WR FRUUHFW IRU VDPSOH VL]H E\ PXOWLSO\LQJ WKH VWDWLVWLF E\

n

n−1 ZKHUHn LV WKH QXPEHU RI JHQH FRSLHV LQ WKH VDPSOH 7KLV LV WKH XQELDVHG

HVWLPDWRU RI KHWHUR]\JRVLW\ 1HL DQG 5R\FKRXGKXU\  DQG LW LV XVHG IRU H[DPSOH LQ 3DSHU ,,,

,Q WKH LQILQLWH VLWHV PRGHO DQG IRU D SRSXODWLRQ DW HTXLOLEULXP H[SHFWHG KHW

HUR]\JRVLW\ LV HTXDO WRΘ WKHWD  WKH ³SRSXODWLRQ PXWDWLRQ SDUDPHWHU´ Θ UHS

UHVHQWV WKH GLYHUVLW\ LQ D SRSXODWLRQ DW PXWDWLRQGULIW HTXLOLEULXP LW LV D IXQF

WLRQ RI WKH HIIHFWLYH SRSXODWLRQ VL]HNeDQG RI WKH PXWDWLRQ UDWHμ Θ = 4Neμ

IRU GLSORLG ORFL  ([SHFWHG KHWHUR]\JRVLW\ LV DQ HVWLPDWRU RIΘ LH D TXDQWLW\

PHDVXUHG IURP D VDPSOH WR HVWLPDWH D SDUDPHWHU WKDW FDQQRW EH PHDVXUHG GL

UHFWO\ ,I ZH KDYH DQ HVWLPDWH RI WKH PXWDWLRQ UDWH DQG RI Θ LW LV SRVVLEOH WR

HVWLPDWH WKH HIIHFWLYH SRSXODWLRQ VL]H

π RU WKH PHDQ QXPEHU RI SDLUZLVH GLIIHUHQFHV RU WKH QXFOHRWLGH GLYHU VLW\ LV DQRWKHU HVWLPDWRU RIΘ 7R FDOFXODWH ʌ LQ D VDPSOH RQH KDV WR FRPSDUH

VHTXHQFHV SDLUZLVH DQG FRXQW WKH QXPEHU RI GLIIHUHQFHV EHWZHHQ WKH VHTXHQFHV LQ HDFK SDLU

7KH VLWH IUHTXHQF\ VSHFWUXP 6)6 RU DOOHOH IUHTXHQF\ VSHFWUXP LV D VH

ULHV RI VXPPDU\ VWDWLVWLFV n − 1 VXPPDU\ VWDWLVWLFV IRU WKH XQIROGHG 6)6 DQG

DSSUR[LPDWHO\ n2 IRU WKH IROGHG 6)6 ZKHUHn LV WKH QXPEHU RI JHQH FRSLHV LQ

(27)

WKH VDPSOH  ,Q WKH XQIROGHG 6)6 ZKLFK UHTXLUHV NQRZOHGJH RI WKH DQFHVWUDO VWDWH RI WKH YDULDQWV  HDFK VWDWLVWLF LV WKH FRXQW RU WKH SURSRUWLRQ RI ORFL WKDW

KDYHi GHULYHG DOOHOHV DQG n−i DQFHVWUDO DOOHOHV  7KH IROGHG 6)6 FDQ EH FDOFX

ODWHG RQ GDWD IRU ZKLFK WKH DQFHVWUDO VWDWH LV QRW NQRZQ LQ WKDW FDVH HDFK FODVV UHSUHVHQWV WKH FRXQW RU SURSRUWLRQ RI ORFL ZLWK D JLYHQ PLQRU DOOHOH IUHTXHQF\

IURP n1 WR12  7KH H[SHFWHG 6)6 XQGHU WKH LGHDO :ULJKW)LVKHU SRSXODWLRQ IRU

D JLYHQ YDOXH RIn LV NQRZQ DQG GHYLDWLRQV IURP LW FDQ EH XVHG WR IRUPXODWH

K\SRWKHVHV UHJDUGLQJ GHPRJUDSKLF RU VHOHFWLYH SURFHVVHV 7KH MRLQW 6)6 FDQ EH FDOFXODWHG IRU WZR SRSXODWLRQV DQG XVHG WR LQIHU GHPRJUDSKLF SDUDPHWHUV ([FRIILHU HW DO   7KH FRQGLWLRQDO 6)6 LV DQRWKHU GHULYDWLYH RI WKH 6)6 ZKHUH WKH 6)6 LV FDOFXODWHG EDVHG RQ RQO\ VRPH YDULDQWV RI D SRSXODWLRQ IRU H[DPSOH WKH YDULDQWV WKDW KDYH WKH GHULYHG DOOHOH LQ DQRWKHU SRSXODWLRQ 'XU YDVXOD DQG 6DQNDUDUDPDQ  

7DMLPD¶V ' DT LV D VXPPDU\ RI WKH 6)6 DQG DQ LQGLFDWRU RI QHXWUDOLW\ ,WV IRUPXOD LV DT = π − S an  V ar(π − S an) 

ZKHUHπ LV WKH VWDWLVWLFV GHVFULEHG SUHYLRXVO\ S LV WKH QXPEHU RI VHJUHJDWLQJ

VLWHV DQGan LV WKH VXP RI WKH VHTXHQFH1 + 12 + 13 + ... + n−11 ZLWKn WKH

QXPEHU RI JHQH FRSLHV LQ WKH VDPSOH S

an RU :DWWHUVRQ¶V Θ LV \HW DQRWKHU

HVWLPDWRU RI Θ :DWWHUVRQ   ,Q DQ LGHDO HJ D :) SRSXODWLRQ XQGHU

WKH LQILQLWH VLWHV PRGHO WKH WZR HVWLPDWRUV DUH HTXDO DQG WKH YDOXH RIDT LV

]HUR 'HYLDWLRQV IURP ]HUR VXJJHVW QRQ QHXWUDO SURFHVVHV RU D GHPRJUDSKLF KLVWRU\ WKDW LV QRW ZHOO DSSUR[LPDWHG E\ DQ LGHDO SRSXODWLRQ PRGHO $ SRVLWLYH

YDOXH RI DT LQGLFDWHV DQ H[FHVV RI LQWHUPHGLDWH IUHTXHQF\ YDULDQWV ZKLFK

FRQWULEXWH PRUH WRπ WKDQ WR S  WKDW FDQ EH GXH WR D SRSXODWLRQ VL]H FRQWUDFWLRQ

SRSXODWLRQ VWUXFWXUH RU EDODQFLQJ VHOHFWLRQ $ QHJDWLYH YDOXH RIDT LQGLFDWHV

DQ H[FHVV RI UDUH YDULDQWV ZKLFK FRQWULEXWH PRUH WR S DQG FDQ EH GXH WR D

SRSXODWLRQ VL]H H[SDQVLRQ RU SRVLWLYH VHOHFWLRQ

FST LV LQ FRQWUDVW WR WKH VWDWLVWLFV GHVFULEHG VR IDU DQ LQWHUSRSXODWLRQ VXP

PDU\ VWDWLVWLF LW LV DQ LQGH[ RI JHQHWLF GLIIHUHQWLDWLRQ EHWZHHQ SRSXODWLRQV ,WV

PRVW H[WUHPH YDOXHV FRUUHVSRQG WR SDQPL[LD FST = 0 WKH SRSXODWLRQV DUH

PDWLQJ UDQGRPO\ DQG WR FRPSOHWH JHQHWLF LVRODWLRQ FST = 1  ,W LV FDOFXODWHG

EHWZHHQ WZR RU PRUH SRSXODWLRQV E\ FRPSDULQJ WKH H[SHFWHG KHWHUR]\JRVLW\ IRU WKH SRSXODWLRQV WDNHQ VHSDUDWHO\ WR WKH H[SHFWHG KHWHUR]\JRVLW\ DVVXPLQJ WKDW WKHVH WZR SRSXODWLRQV ZHUH LQ IDFW D VLQJOH ZHOOPL[HG RU ³SDQPLFWLF´ SRSXODWLRQ ,W LV EDVHG RQ DOOHOH IUHTXHQFLHV LQ WKH SRSXODWLRQV D SUDFWLFDOO\ VSHDNLQJ XQREVHUYDEOH TXDQWLW\ EXW WKDW FDQ EH HVWLPDWHG IURP D SRSXODWLRQ VDPSOH /DUJHU VDPSOH VL]HV ZLOO JLYH PRUH DFFXUDWH UHVXOWV

, GLVFXVV RQH PRUH FDWHJRU\ RI VXPPDU\ VWDWLVWLFV ZKLFK LV EDVHG RQ VWUHWFKHV RI YDULDQWV EXW GR QRW UHTXLUH NQRZOHGJH RI WKH SKDVH UXQV RI KRPR]\JRVLW\ 52+ %URPDQ DQG :HEHU  .LULQ HW DO   52+ DQDO\VLV EXLOGV RQ 

(28)

LGHQWLI\LQJ VWUHWFKHV LQ WKH JHQRPH RI DQ LQGLYLGXDO ZKHUH DOO SRVLWLRQV DUH KR PR]\JRXV :KLOH VKRUW 52+ FDQ EH GXH WR FKDQFH ORQJHU VWUHWFKHV DUH PRUH OLNHO\ WR UHIOHFW SKHQRPHQD VXFK DV LQEUHHGLQJ DQGRU VPDOO SRSXODWLRQ VL]H 7KH GHILQLWLRQ RI 52+ ± IRU H[DPSOH WKH PLQLPDO OHQJWK RI D VWUHWFK WR TXDOLI\ DV 52+ RU WKH QXPEHU RI ³PLVWDNHV´ WKDW RQH DOORZV ± GHSHQGV RQ WKH W\SH RI GDWD XVHG ,Q 3DSHUV , ,,, DQG ,9 , XVHG GLIIHUHQW SDUDPHWHUV GHSHQGLQJ RQ WKH W\SH RI GDWD

:KLOH HDFK VXPPDU\ VWDWLVWLFV KDV D SUHFLVH GHILQLWLRQ WKHUH PD\ EH YDULRXV ZD\V DQG IRUPXOD WR HVWLPDWH HDFK RI WKHP IURP UHDO GDWD DQG LW LV LPSRUWDQW WR NHHS WUDFN RI WKH PHWKRGV XVHG LQ D JLYHQ VWXG\ ,Q WKLV WKHVLV VXPPDU\ VWDWLV WLFV KDYH EHHQ FRPSXWHG ZLWK FXVWRP VFULSWV ZLWK WKH VRIWZDUH SOLQN 3XUFHOO HW DO   RU ZLWK 3\WKRQ VFULSWV IROORZLQJ -D\ HW DO  DQG UHIHUHQFHV WKHUHLQ

 9LVXDO H[SORUDWLRQ RI JHQHWLF GDWD

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¶V E\ /XLJL /XFD &DYDOOL6IRU]D LQ KLV ERRN ³7KH KLVWRU\ DQG JHRJUD SK\ RI KXPDQ JHQHV´ &DYDOOL6IRU]D HW DO   0RUH UHFHQWO\ 1RYHPEUH HW DO  VKRZHG WKDW WKH SURMHFWLRQV RI WKH PDLQ D[HV RI YDULDWLRQ LQ D VHW RI JHQHWLF PDUNHUV FRUUHODWHG ZLWK JHRJUDSK\ IRU (XURSHDQ SRSXODWLRQV $Q FLHQW VDPSOHV FDQ EH DGGHG WR 3& DQDO\VHV WKH FRPPRQ SURFHGXUH LQ WKDW FDVH LV WR XVH UHIHUHQFH PRGHUQ VDPSOHV WR GHILQH WKH 3& D[LV DQG WR ³SURMHFW´ WKH DQFLHQW VDPSOHV LQWR WKH PRGHUQ GLYHUVLW\

$Q DOWHUQDWLYH DSSURDFK WR 3&$ LV WR GR D 0XOWLGLPHQVLRQDO 6FDOLQJ 3ORW 0'6 EDVHG RQ D TXDQWLW\ RI LQWHUHVW IRU H[DPSOH SDLUZLVH DOOHOH VKDULQJ

GLVWDQFHV RUFST 'HSHQGLQJ RQ WKH FKRVHQ TXDQWLW\ WKLV PHWKRG FDQ EH YHU\

IDVW 0RUHRYHU RQFH DQ LQWHUHVWLQJ SDWWHUQ LV REVHUYHG LW LV SRVVLEOH WR JR EDFN WR WKH ELRORJLFDOO\ PHDQLQJIXO VWDWLVWLFV ZKLFK LV QRW DV VWUDLJKWIRUZDUG IRU WKH YHFWRUV XQGHUO\LQJ 3&$ D[HV 

(29)

 &OXVWHULQJ PHWKRGV

0RGHOEDVHG FOXVWHULQJ PHWKRGV WKDW LQIHU FOXVWHUV PD[LPL]LQJ DOOHOH IUHTXHQ FLHV GLIIHUHQFHV OLNH $'0,;785( $OH[DQGHU HW DO  DQG 6758& 785( 3ULWFKDUG HW DO   DUH DOVR YHU\ SRSXODU LQ KXPDQ SRSXODWLRQ JH QHWLFV %RWK LQIHU JHQHWLF PHPEHUVKLSV IRU LQGLYLGXDOV WR SUHGHILQHG QXPEHUV RI YLUWXDO FOXVWHUV UHIHUUHG WR DV ³.´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

RI DGPL[WXUH HJfVWDWLVWLFV :HLU DQG &RFNHUKDP  :ULJKW  

 ,QIHUULQJ GHPRJUDSKLF KLVWRU\

9DULRXV SURFHVVHV VKDSH WKH GHPRJUDSKLF KLVWRU\ RI SRSXODWLRQV DQG VSHFLHV 6RPH GHPRJUDSKLF SURFHVVHV OLNH SRSXODWLRQ VSOLWV DQG LVRODWLRQE\GLVWDQFH FRQWULEXWH WR VHSDUDWH OLQHDJHV RWKHUV EULQJ WKHP WRJHWKHU IRU H[DPSOH WKURXJK DGPL[WXUH HYHQWV RU UHSHDWHG PLJUDWLRQV $QRWKHU DVSHFW RI GHPRJUDSK\ DUH FKDQJHV LQ SRSXODWLRQ VL]H

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

PRGHOV *LYHQ REVHUYHG GDWDD DQG D SDUDPHWHU RI LQWHUHVW Θ ZH DUH LQWHU

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

(30)

DOORZ HIILFLHQW H[SORUDWLRQ RI WKH SDUDPHWHU VSDFH %D\HVLDQ DSSURDFKHV WDNH D GLIIHUHQW DQJOH ZLWK SDUDPHWHUV KDYLQJ D SUREDELOLW\ GLVWULEXWLRQ LQ 0/ DS SURDFKHV SDUDPHWHUV KDYH D ILQLWH YDOXH  %DVHG RQ SULRU NQRZOHGJH RQH GH ILQHV D SULRU GLVWULEXWLRQ IRU WKH SDUDPHWHU IRU H[DPSOH IRU SRSXODWLRQ VL]HV WKH SULRU GLVWULEXWLRQ FRXOG EH WKH UDQJH RI VL]HV REVHUYHG LQ QDWXUDO SRSXOD WLRQV  7KHQ RQH JDWKHUV GDWD RU SHUIRUPV DQ H[SHULPHQW DQG EDVHG RQ WKDW REWDLQV WKH SRVWHULRU GLVWULEXWLRQ RI WKH SDUDPHWHU LH IRU D UDQJH RI YDOXHV WKH SUREDELOLW\ WKDW WKH SDUDPHWHU WDNHV D YDOXH JLYHQ WKH GDWD 

7KH IROORZLQJ VHFWLRQV SUHVHQW GLIIHUHQW GHPRJUDSKLF SURFHVVHV DQG PHWK RGV WKDW DLP WR FKDUDFWHULVH WKHP 0RVW RI WKHVH PHWKRGV DUH EDVHG RQ 0/ RU RQ %D\HVLDQ DSSURDFKHV ,Q WKH HQG , LQWURGXFH D IOH[LEOH IUDPHZRUN $S SUR[LPDWH %D\HVLDQ &RPSXWDWLRQ WKDW FDQ EH DSSOLHG WR WHVW DQ\ GHPRJUDSKLF SURFHVV DJDLQVW D VHW RI REVHUYHG GDWD E\ JHQHUDWLQJ VLPXODWHG GDWD XQGHU SRV VLEOH VFHQDULRV 7KLV FDQ WKHQ EH XVHG WR FKRRVH EHWZHHQ VFHQDULRV DQG WR HVWLPDWH SDUDPHWHUV VHH 3DSHU ,,, 

 'LYHUJHQFH

³'LYHUJHQFH´ RU ³VSOLW´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³FOHDQ´ LQWR WZR RU PRUH JURXSV IURP RQH JHQHUDWLRQ WR WKH QH[W ZLWKRXW DQ\ VXEVHTXHQW FRQWDFW EHWZHHQ WKH SRSXODWLRQV HYHQ LI WKDW LV RIWHQ DVVXPHG LQ PRGHOV 2IWHQ GL YHUJHQFH LV SURJUHVVLYH LW PLJKW VWDUW ZLWK SRSXODWLRQ VWUXFWXUH LH ZLWKLQ D SRSXODWLRQ LQGLYLGXDOV UHSURGXFH SUHIHUDEO\ ZLWK VRPH LQGLYLGXDOV ,W LV DOVR FRPPRQ WKDW GLYHUJLQJ SRSXODWLRQV FRQWLQXH WR H[FKDQJH PLJUDQWV DIWHU WKH VSOLW

2QH ZD\ WR FKDUDFWHULVH GLYHUJHQFH LV WR HVWLPDWH WKH WLPH WR WKH PRVW UH

FHQW FRPPRQ DQFHVWRU 705&$  LH WKH \RXQJHVW LQGLYLGXDO ZKR LV DQ DQ

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

References

Related documents

Wkh sxusrvh ri wklv uhsruw lv wr ghvfuleh wkh prgho wkdw wkh Vzhglvk Qdwlrqdo Ghew R!fh kdv ghyhorshg iru vwxg|lqj wkh hhfwv ri wkh fkrlfh ri gxudwlrq dqg fxuuhqf| frpsrvlwlrq lq

'HWHUPLQDWLRQ RI WKH SDUWLDO FRHIILFLHQW ηγ P  ZLWK WKH WDEOH PHWKRG DQG WKH SUREDELOLVWLFPHWKRGJLYHVVOLJKWO\GLIIHUHQWUHVXOWVIRUWKHVHYHQEULGJHVEHWZHHQ /XOHn

VRIWZDUH DQG WHFKQLTXHV WKH LQIRUPDWLRQ VKRXOG EH SUHVHQWHG WR WKH SXEOLF LQ DQ HDV\WRXVH DQG HDV\WR XQGHUVWDQG ZD\ LQ RUGHU WR UHDFK WKH ODUJHVW SRVVLEOH SHUFHQWDJH RI WKH WDUJHW

DQG DVVXPSWLRQV WKDW DIIHFW WKH UHSRUWHG DPRXQWV RI DVVHWV OLDELOLWLHV DQG H[SHQGLWXUHV 7KHVH HVWLPDWHV DQG. DVVXPSWLRQV DUH EDVHG RQ PDQDJHPHQW¶V

Lq ghyhorslqj wkh vlpxodwlrq prgho/rxu dpelwlrq kdv ehhq wr nhhs wkh prgho vlpsoh dqg wr plqlpl}h wkh qxpehu ri yduldeohv wr eh prghoohg1 Frq0 vhtxhqwo|/wkh prgho jlyhv d

Dv fdq eh vhhq iurp Iljxuh  dqg Iljxuh 2 wkh htxdwlrq E  jlyhv d eldvhg uhvxow iru qlwh vdpsoh vl}hv1 Orrnlqj edfn dw wkh ghulydwlyhv ri htxdwlrq E  lw fdq eh vhhq wkdw

Rughu vwdwlvwlfv edvhg hvwlpdwruv duh riwhq xvhg iru vpdoo vdpsohv zkhuh dv|pswrwlf wkhru| lv lpsurshu/ dqg lw lv wkhuhiruh lpsruwdqw wr nqrz wkh vpdoo vdpsoh surshuwlhv dqg

 2SWLPLVDWLRQ RI WKH SURWHFWLRQ HTXLSPHQW GHSHQGLQJ RQ WKH YROWDJH VDJ  PDJQLWXGH DQG GXUDWLRQ  IRU D V\QFKURQRXV PDFKLQH WR NHHS V\QFKURQLVP. ZKHQ