Asymptotic theory of rerandomization in treatment-control experiments

Although complete randomization ensures covariate balance on average, the chance of observing significant differences between treatment and control covariate distributions increases with many covariates. Rerandomization discards randomizations that do not satisfy a predetermined covariate balance cr...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Proceedings of the National Academy of Sciences - PNAS Ročník 115; číslo 37; s. 9157
Hlavní autori: Li, Xinran, Ding, Peng, Rubin, Donald B
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: United States 11.09.2018
Predmet:
ISSN:1091-6490, 1091-6490
On-line prístup:Zistit podrobnosti o prístupe
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Although complete randomization ensures covariate balance on average, the chance of observing significant differences between treatment and control covariate distributions increases with many covariates. Rerandomization discards randomizations that do not satisfy a predetermined covariate balance criterion, generally resulting in better covariate balance and more precise estimates of causal effects. Previous theory has derived finite sample theory for rerandomization under the assumptions of equal treatment group sizes, Gaussian covariate and outcome distributions, or additive causal effects, but not for the general sampling distribution of the difference-in-means estimator for the average causal effect. We develop asymptotic theory for rerandomization without these assumptions, which reveals a non-Gaussian asymptotic distribution for this estimator, specifically a linear combination of a Gaussian random variable and truncated Gaussian random variables. This distribution follows because rerandomization affects only the projection of potential outcomes onto the covariate space but does not affect the corresponding orthogonal residuals. We demonstrate that, compared with complete randomization, rerandomization reduces the asymptotic quantile ranges of the difference-in-means estimator. Moreover, our work constructs accurate large-sample confidence intervals for the average causal effect.
AbstractList Although complete randomization ensures covariate balance on average, the chance of observing significant differences between treatment and control covariate distributions increases with many covariates. Rerandomization discards randomizations that do not satisfy a predetermined covariate balance criterion, generally resulting in better covariate balance and more precise estimates of causal effects. Previous theory has derived finite sample theory for rerandomization under the assumptions of equal treatment group sizes, Gaussian covariate and outcome distributions, or additive causal effects, but not for the general sampling distribution of the difference-in-means estimator for the average causal effect. We develop asymptotic theory for rerandomization without these assumptions, which reveals a non-Gaussian asymptotic distribution for this estimator, specifically a linear combination of a Gaussian random variable and truncated Gaussian random variables. This distribution follows because rerandomization affects only the projection of potential outcomes onto the covariate space but does not affect the corresponding orthogonal residuals. We demonstrate that, compared with complete randomization, rerandomization reduces the asymptotic quantile ranges of the difference-in-means estimator. Moreover, our work constructs accurate large-sample confidence intervals for the average causal effect.Although complete randomization ensures covariate balance on average, the chance of observing significant differences between treatment and control covariate distributions increases with many covariates. Rerandomization discards randomizations that do not satisfy a predetermined covariate balance criterion, generally resulting in better covariate balance and more precise estimates of causal effects. Previous theory has derived finite sample theory for rerandomization under the assumptions of equal treatment group sizes, Gaussian covariate and outcome distributions, or additive causal effects, but not for the general sampling distribution of the difference-in-means estimator for the average causal effect. We develop asymptotic theory for rerandomization without these assumptions, which reveals a non-Gaussian asymptotic distribution for this estimator, specifically a linear combination of a Gaussian random variable and truncated Gaussian random variables. This distribution follows because rerandomization affects only the projection of potential outcomes onto the covariate space but does not affect the corresponding orthogonal residuals. We demonstrate that, compared with complete randomization, rerandomization reduces the asymptotic quantile ranges of the difference-in-means estimator. Moreover, our work constructs accurate large-sample confidence intervals for the average causal effect.
Although complete randomization ensures covariate balance on average, the chance of observing significant differences between treatment and control covariate distributions increases with many covariates. Rerandomization discards randomizations that do not satisfy a predetermined covariate balance criterion, generally resulting in better covariate balance and more precise estimates of causal effects. Previous theory has derived finite sample theory for rerandomization under the assumptions of equal treatment group sizes, Gaussian covariate and outcome distributions, or additive causal effects, but not for the general sampling distribution of the difference-in-means estimator for the average causal effect. We develop asymptotic theory for rerandomization without these assumptions, which reveals a non-Gaussian asymptotic distribution for this estimator, specifically a linear combination of a Gaussian random variable and truncated Gaussian random variables. This distribution follows because rerandomization affects only the projection of potential outcomes onto the covariate space but does not affect the corresponding orthogonal residuals. We demonstrate that, compared with complete randomization, rerandomization reduces the asymptotic quantile ranges of the difference-in-means estimator. Moreover, our work constructs accurate large-sample confidence intervals for the average causal effect.
Author Ding, Peng
Rubin, Donald B
Li, Xinran
Author_xml – sequence: 1
  givenname: Xinran
  surname: Li
  fullname: Li, Xinran
  organization: Department of Statistics, Harvard University, Cambridge, MA 02138
– sequence: 2
  givenname: Peng
  surname: Ding
  fullname: Ding, Peng
  organization: Department of Statistics, University of California, Berkeley, CA 94720
– sequence: 3
  givenname: Donald B
  surname: Rubin
  fullname: Rubin, Donald B
  email: dbrubin@mac.com
  organization: Department of Statistics, Harvard University, Cambridge, MA 02138; dbrubin@mac.com
BackLink https://www.ncbi.nlm.nih.gov/pubmed/30150408$$D View this record in MEDLINE/PubMed
BookMark eNpNj81LxDAUxIOsuB969iY9eun63ibZJsdlWT9gwYueSzb7ipU2qUkK1r_eiit4mmH4MczM2cR5R4xdIywRCn7XOROXqEChRkR5xmYIGvO10DD556dsHuM7AGip4IJNOaAEAWrGdps4tF3yqbZZeiMfhsxXWaBg3NG39ZdJtXdZ7bIUyKSWXMqtdyn4JqPPjkL9E8VLdl6ZJtLVSRfs9X73sn3M988PT9vNPrei0CnXAgvkSpGQlRIVqEJCtQYuwBy1AQQiIVBJroyyRlXSWiJ5IIO80MLy1YLd_vZ2wX_0FFPZ1tFS0xhHvo_lajwoOQCqEb05of2hpWPZjVNNGMq_66tv6YddkA
CitedBy_id crossref_primary_10_1093_biomet_asz026
crossref_primary_10_1002_sim_9508
crossref_primary_10_1111_biom_13712
crossref_primary_10_1093_jrsssb_qkad080
crossref_primary_10_1080_00031305_2020_1717619
crossref_primary_10_1080_01621459_2025_2507432
crossref_primary_10_1126_science_adr7394
crossref_primary_10_1016_j_jeconom_2025_106049
crossref_primary_10_1016_j_jspi_2021_09_002
crossref_primary_10_1111_insr_12498
crossref_primary_10_1002_cjs_11765
crossref_primary_10_1214_22_AOS2235
crossref_primary_10_1002_cjs_11685
crossref_primary_10_1080_07350015_2024_2403381
crossref_primary_10_1515_scid_2024_0001
crossref_primary_10_1111_rssb_12417
crossref_primary_10_1093_ectj_utab009
crossref_primary_10_3390_e26121023
crossref_primary_10_1016_j_jspi_2021_11_005
crossref_primary_10_1093_biomet_asaa103
crossref_primary_10_1080_01621459_2022_2102985
crossref_primary_10_1093_biomet_asac045
crossref_primary_10_1093_jrsssb_qkaf002
crossref_primary_10_1016_j_jspi_2024_106241
crossref_primary_10_1111_rssc_12513
crossref_primary_10_1080_01621459_2022_2049278
crossref_primary_10_1080_10618600_2020_1753531
crossref_primary_10_3102_10769986211027240
crossref_primary_10_1093_biomet_asad027
crossref_primary_10_1002_cjs_11783
crossref_primary_10_1093_jrsssb_qkad017
crossref_primary_10_1093_biomtc_ujaf036
crossref_primary_10_1111_rssb_12439
crossref_primary_10_1080_01621459_2024_2412363
crossref_primary_10_1002_sta4_70058
crossref_primary_10_1186_s12874_023_01977_7
crossref_primary_10_1016_j_jspi_2020_07_002
crossref_primary_10_1214_18_AOS1790
crossref_primary_10_1080_01621459_2020_1863221
crossref_primary_10_1080_01621459_2021_1984926
crossref_primary_10_1214_25_AOS2513
crossref_primary_10_1080_01621459_2023_2285474
crossref_primary_10_1016_j_jeconom_2024_105724
crossref_primary_10_1080_01621459_2024_2366043
crossref_primary_10_1093_ectj_utz020
crossref_primary_10_1111_rssb_12353
crossref_primary_10_1080_07350015_2024_2429464
crossref_primary_10_1016_j_csda_2022_107642
crossref_primary_10_1080_01621459_2021_1990767
crossref_primary_10_1080_19466315_2023_2267774
ContentType Journal Article
DBID CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1073/pnas.1808191115
DatabaseName Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList MEDLINE - Academic
MEDLINE
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod no_fulltext_linktorsrc
Discipline Sciences (General)
EISSN 1091-6490
ExternalDocumentID 30150408
Genre Research Support, U.S. Gov't, Non-P.H.S
Research Support, Non-U.S. Gov't
Journal Article
Research Support, N.I.H., Extramural
GrantInformation_xml – fundername: NIAID NIH HHS
  grantid: R01 AI102710
GroupedDBID ---
-DZ
-~X
.55
0R~
123
29P
2AX
2FS
2WC
4.4
53G
5RE
5VS
85S
AACGO
AAFWJ
AANCE
ABBHK
ABOCM
ABPLY
ABPPZ
ABTLG
ABXSQ
ABZEH
ACGOD
ACHIC
ACIWK
ACNCT
ACPRK
ADQXQ
ADULT
AENEX
AEUPB
AEXZC
AFFNX
AFOSN
AFRAH
ALMA_UNASSIGNED_HOLDINGS
AQVQM
BKOMP
CGR
CS3
CUY
CVF
D0L
DCCCD
DIK
DU5
E3Z
EBS
ECM
EIF
EJD
F5P
FRP
GX1
H13
HH5
HYE
IPSME
JAAYA
JBMMH
JENOY
JHFFW
JKQEH
JLS
JLXEF
JPM
JSG
JST
KQ8
L7B
LU7
N9A
NPM
N~3
O9-
OK1
PNE
PQQKQ
R.V
RHI
RNA
RNS
RPM
RXW
SA0
SJN
TAE
TN5
UKR
W8F
WH7
WOQ
WOW
X7M
XSW
Y6R
YBH
YKV
YSK
ZCA
~02
~KM
7X8
ID FETCH-LOGICAL-c479t-94171388e45f84f08750f60340ad9a010ee4418538a8ca8f5ccee5bea13794c32
IEDL.DBID 7X8
ISICitedReferencesCount 63
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000444257200049&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1091-6490
IngestDate Fri Sep 05 06:26:34 EDT 2025
Thu Apr 03 07:05:21 EDT 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 37
Keywords quantile range
causal inference
geometry of rerandomization
Mahalanobis distance
covariate balance
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c479t-94171388e45f84f08750f60340ad9a010ee4418538a8ca8f5ccee5bea13794c32
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
OpenAccessLink http://nrs.harvard.edu/urn-3:HUL.InstRepos:42615132
PMID 30150408
PQID 2095530018
PQPubID 23479
ParticipantIDs proquest_miscellaneous_2095530018
pubmed_primary_30150408
PublicationCentury 2000
PublicationDate 2018-09-11
PublicationDateYYYYMMDD 2018-09-11
PublicationDate_xml – month: 09
  year: 2018
  text: 2018-09-11
  day: 11
PublicationDecade 2010
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle Proceedings of the National Academy of Sciences - PNAS
PublicationTitleAlternate Proc Natl Acad Sci U S A
PublicationYear 2018
SSID ssj0009580
Score 2.5973768
Snippet Although complete randomization ensures covariate balance on average, the chance of observing significant differences between treatment and control covariate...
SourceID proquest
pubmed
SourceType Aggregation Database
Index Database
StartPage 9157
SubjectTerms Models, Theoretical
Random Allocation
Title Asymptotic theory of rerandomization in treatment-control experiments
URI https://www.ncbi.nlm.nih.gov/pubmed/30150408
https://www.proquest.com/docview/2095530018
Volume 115
WOSCitedRecordID wos000444257200049&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV09T8MwELWAMrAA5bN8yUgMMBjs2o7tCVWoFQtVB5C6RY7jSB2ahCYg9d9jJy5lQUJiyeJEis539rvz-T0AbvqEWgd0CUotMYgJY5A2BiPrudKjiPEM60ZsQozHcjpVk1Bwq0Jb5WpNbBbqtDC-Ru6SdOUVbjCRj-U78qpR_nQ1SGhsgg51UMYHppjKH6S7smUjUARFTOEVtY-gD2Wuq3viZSd8tPPf8WWzz4z2_vuH-2A3IEw4aF2iCzZsfgC6IYYreBuIpu8OwXBQLedlXbgXYXOhcQmLDC7caJ4W83BBE85y-N2NjkJnO1wLA1RH4G00fH16RkFWARkmVI0UIy4zldK6iZAs85T2OIswZVinSrv8zFrmKW2o1NJomXHjNlKeWE2oC15D-8dgKy9yewpgpAjXPGGJIIbZ1KhMKJww7WAal8LwHrhemSp2buvPInRui48qXhurB05ae8dly68RU1-FYVie_eHrc7DjIEzTwUHIBehkLmjtJdg2n_WsWlw1_uCe48nLF0l9v6M
linkProvider ProQuest
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Asymptotic+theory+of+rerandomization+in+treatment-control+experiments&rft.jtitle=Proceedings+of+the+National+Academy+of+Sciences+-+PNAS&rft.au=Li%2C+Xinran&rft.au=Ding%2C+Peng&rft.au=Rubin%2C+Donald+B&rft.date=2018-09-11&rft.issn=1091-6490&rft.eissn=1091-6490&rft.volume=115&rft.issue=37&rft.spage=9157&rft_id=info:doi/10.1073%2Fpnas.1808191115&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1091-6490&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1091-6490&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1091-6490&client=summon