Improving massive experiments with threshold blocking

↓ Direkt till sidans innehåll
↓ Direkt till sidans sekundära innehåll (sidomenyn)

Sökning: WFRF:(Higgins S.) > (2015-2019) > Improving massive e...

LIBRIS Formathandbok (Information om MARC21)

Fältnamn	Indikatorer	Metadata
000		03385naa a2200361 4500
001		oai:DiVA.org:uu-299820
003		SwePub
008		160728s2016 \| \|\|\|\|\|\|\|\|\|\|\|000 \|\|eng\|
024	7	a https://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-2998202 URI
024	7	a https://doi.org/10.1073/pnas.15105041132 DOI
040		a (SwePub)uu
041		a engb eng
042		9 SwePub
072	7	a ref2 swepub-contenttype
072	7	a art2 swepub-publicationtype
100	1	a Higgins, Michael J.u Kansas State Univ, Dept Stat, Manhattan, KS 66506 USA4 aut
245	1 0	a Improving massive experiments with threshold blocking
264		c 2016-07-05
264	1	b Proceedings of the National Academy of Sciences,c 2016
338		a print2 rdacarrier
520		a Inferences from randomized experiments can be improved by blocking: assigning treatment in fixed proportions within groups of similar units. However, the use of the method is limited by the difficulty in deriving these groups. Current blocking methods are restricted to special cases or run in exponential time; are not sensitive to clustering of data points; and are often heuristic, providing an unsatisfactory solution in many common instances. We present an algorithm that implements a widely applicable class of blocking-threshold blocking-that solves these problems. Given a minimum required group size and a distance metric, we study the blocking problem of minimizing the maximum distance between any two units within the same group. We prove this is a nondeterministic polynomial-time hard problem and derive an approximation algorithm that yields a blocking where the maximum distance is guaranteed to be, at most, four times the optimal value. This algorithm runs in O(n log n) time with O(n) space complexity. This makes it, to our knowledge, the first blocking method with an ensured level of performance that works in massive experiments. Whereas many commonly used algorithms form pairs of units, our algorithm constructs the groups flexibly for any chosen minimum size. This facilitates complex experiments with several treatment arms and clustered data. A simulation study demonstrates the efficiency and efficacy of the algorithm; tens of millions of units can be blocked using a desktop computer in a few minutes.
650	7	a SAMHÄLLSVETENSKAPx Ekonomi och näringslivx Nationalekonomi0 (SwePub)502012 hsv//swe
650	7	a SOCIAL SCIENCESx Economics and Businessx Economics0 (SwePub)502012 hsv//eng
653		a experimental design; blocking; big data; causal inference
653		a Nationalekonomi
653		a Economics
700	1	a Sävje, Fredriku Uppsala universitet,Nationalekonomiska institutionen4 aut0 (Swepub:uu)fresa529
700	1	a Sekhon, Jasjeet S.u Univ Calif Berkeley, Dept Polit Sci, Berkeley, CA 94720 USA;Univ Calif Berkeley, Dept Stat, Berkeley, CA 94720 USA4 aut
710	2	a Kansas State Univ, Dept Stat, Manhattan, KS 66506 USAb Nationalekonomiska institutionen4 org
773	0	t Proceedings of the National Academy of Sciences of the United States of Americad : Proceedings of the National Academy of Sciencesg 13:27, s. 7369-7376q 13:27<7369-7376x 0027-8424x 1091-6490
856	4	u https://europepmc.org/articles/pmc4941468?pdf=render
856	4 8	u https://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-299820
856	4 8	u https://doi.org/10.1073/pnas.1510504113

Hitta via bibliotek

Proceedings of the National Academy of Sciences of the United States of America (Sök värdpublikationen i LIBRIS)

Till lärosätets databas

Hitta mer i SwePub

Av författaren/redakt...: Higgins, Michael ...; Sävje, Fredrik; Sekhon, Jasjeet ...

Om ämnet

SAMHÄLLSVETENSKAP: SAMHÄLLSVETENSKA ...; och Ekonomi och näri ...; och Nationalekonomi

Artiklar i publikationen: Proceedings of t ...

Av lärosätet: Uppsala universitet

Sök utanför SwePub

Sök vidare i:: Google; Google Book Search; Google Scholar

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.

LIBRIS.kb.se