Strategy-proof school choice mechanisms with minimum quotas and initial endowments

Naoto Hamada, Chia Ling Hsu, Ryoji Kurata, Takamasa Suzuki, Suguru Ueda, Makoto Yokoo

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

We consider a school choice program where minimum quotas are imposed for each school, i.e., a school must be assigned at least a certain number of students to operate. We require that the obtained matching must respect the initial endowments, i.e., each student must be assigned to a school that is at least as good as her initial endowment school. Although minimum quotas are relevant in school choice programs and strategy-proofness is important to many policymakers, few existing mechanisms simultaneously achieve both. One difficulty is that no strategy-proof mechanism exists that is both efficient and fair under the presence of minimum quotas. Furthermore, existing mechanisms require that all students consider all schools acceptable to obtain a feasible matching that respects minimum quotas. This assumption is unrealistic in a school choice program. We consider the environment where a student considers her initial endowment school acceptable and the initial endowments satisfy all the minimum quotas. We develop two strategy-proof mechanisms. One mechanism, which we call the Top Trading Cycles among Representatives with Supplementary Seats (TTCR-SS), is based on the Top Trading Cycles (TTC) mechanism and is significantly extended to handle the supplementary seats of schools while respecting minimum quotas. TTCR-SS is Pareto efficient. The other mechanism, which we call Priority List-based Deferred Acceptance with Minimum Quotas (PLDA-MQ), is based on the Deferred Acceptance (DA) mechanism. PLDA-MQ is fair, satisfies a concept called Priority List-based (PL-) stability, and obtains the student-optimal matching within all PL-stable matchings. Our simulation results show that our new mechanisms are significantly better than simple extensions of the existing mechanisms.

Original languageEnglish
Pages (from-to)47-71
Number of pages25
JournalArtificial Intelligence
Volume249
DOIs
Publication statusPublished - Aug 2017

Fingerprint

school choice
Students
Seats
school
acceptance
student
School Choice
respect
simulation

All Science Journal Classification (ASJC) codes

  • Language and Linguistics
  • Linguistics and Language
  • Artificial Intelligence

Cite this

Strategy-proof school choice mechanisms with minimum quotas and initial endowments. / Hamada, Naoto; Hsu, Chia Ling; Kurata, Ryoji; Suzuki, Takamasa; Ueda, Suguru; Yokoo, Makoto.

In: Artificial Intelligence, Vol. 249, 08.2017, p. 47-71.

Research output: Contribution to journalArticle

Hamada, Naoto ; Hsu, Chia Ling ; Kurata, Ryoji ; Suzuki, Takamasa ; Ueda, Suguru ; Yokoo, Makoto. / Strategy-proof school choice mechanisms with minimum quotas and initial endowments. In: Artificial Intelligence. 2017 ; Vol. 249. pp. 47-71.
@article{335b33dcbf7a4703876a6d33ad8d42ed,
title = "Strategy-proof school choice mechanisms with minimum quotas and initial endowments",
abstract = "We consider a school choice program where minimum quotas are imposed for each school, i.e., a school must be assigned at least a certain number of students to operate. We require that the obtained matching must respect the initial endowments, i.e., each student must be assigned to a school that is at least as good as her initial endowment school. Although minimum quotas are relevant in school choice programs and strategy-proofness is important to many policymakers, few existing mechanisms simultaneously achieve both. One difficulty is that no strategy-proof mechanism exists that is both efficient and fair under the presence of minimum quotas. Furthermore, existing mechanisms require that all students consider all schools acceptable to obtain a feasible matching that respects minimum quotas. This assumption is unrealistic in a school choice program. We consider the environment where a student considers her initial endowment school acceptable and the initial endowments satisfy all the minimum quotas. We develop two strategy-proof mechanisms. One mechanism, which we call the Top Trading Cycles among Representatives with Supplementary Seats (TTCR-SS), is based on the Top Trading Cycles (TTC) mechanism and is significantly extended to handle the supplementary seats of schools while respecting minimum quotas. TTCR-SS is Pareto efficient. The other mechanism, which we call Priority List-based Deferred Acceptance with Minimum Quotas (PLDA-MQ), is based on the Deferred Acceptance (DA) mechanism. PLDA-MQ is fair, satisfies a concept called Priority List-based (PL-) stability, and obtains the student-optimal matching within all PL-stable matchings. Our simulation results show that our new mechanisms are significantly better than simple extensions of the existing mechanisms.",
author = "Naoto Hamada and Hsu, {Chia Ling} and Ryoji Kurata and Takamasa Suzuki and Suguru Ueda and Makoto Yokoo",
year = "2017",
month = "8",
doi = "10.1016/j.artint.2017.04.006",
language = "English",
volume = "249",
pages = "47--71",
journal = "Artificial Intelligence",
issn = "0004-3702",
publisher = "Elsevier",

}

TY - JOUR

T1 - Strategy-proof school choice mechanisms with minimum quotas and initial endowments

AU - Hamada, Naoto

AU - Hsu, Chia Ling

AU - Kurata, Ryoji

AU - Suzuki, Takamasa

AU - Ueda, Suguru

AU - Yokoo, Makoto

PY - 2017/8

Y1 - 2017/8

N2 - We consider a school choice program where minimum quotas are imposed for each school, i.e., a school must be assigned at least a certain number of students to operate. We require that the obtained matching must respect the initial endowments, i.e., each student must be assigned to a school that is at least as good as her initial endowment school. Although minimum quotas are relevant in school choice programs and strategy-proofness is important to many policymakers, few existing mechanisms simultaneously achieve both. One difficulty is that no strategy-proof mechanism exists that is both efficient and fair under the presence of minimum quotas. Furthermore, existing mechanisms require that all students consider all schools acceptable to obtain a feasible matching that respects minimum quotas. This assumption is unrealistic in a school choice program. We consider the environment where a student considers her initial endowment school acceptable and the initial endowments satisfy all the minimum quotas. We develop two strategy-proof mechanisms. One mechanism, which we call the Top Trading Cycles among Representatives with Supplementary Seats (TTCR-SS), is based on the Top Trading Cycles (TTC) mechanism and is significantly extended to handle the supplementary seats of schools while respecting minimum quotas. TTCR-SS is Pareto efficient. The other mechanism, which we call Priority List-based Deferred Acceptance with Minimum Quotas (PLDA-MQ), is based on the Deferred Acceptance (DA) mechanism. PLDA-MQ is fair, satisfies a concept called Priority List-based (PL-) stability, and obtains the student-optimal matching within all PL-stable matchings. Our simulation results show that our new mechanisms are significantly better than simple extensions of the existing mechanisms.

AB - We consider a school choice program where minimum quotas are imposed for each school, i.e., a school must be assigned at least a certain number of students to operate. We require that the obtained matching must respect the initial endowments, i.e., each student must be assigned to a school that is at least as good as her initial endowment school. Although minimum quotas are relevant in school choice programs and strategy-proofness is important to many policymakers, few existing mechanisms simultaneously achieve both. One difficulty is that no strategy-proof mechanism exists that is both efficient and fair under the presence of minimum quotas. Furthermore, existing mechanisms require that all students consider all schools acceptable to obtain a feasible matching that respects minimum quotas. This assumption is unrealistic in a school choice program. We consider the environment where a student considers her initial endowment school acceptable and the initial endowments satisfy all the minimum quotas. We develop two strategy-proof mechanisms. One mechanism, which we call the Top Trading Cycles among Representatives with Supplementary Seats (TTCR-SS), is based on the Top Trading Cycles (TTC) mechanism and is significantly extended to handle the supplementary seats of schools while respecting minimum quotas. TTCR-SS is Pareto efficient. The other mechanism, which we call Priority List-based Deferred Acceptance with Minimum Quotas (PLDA-MQ), is based on the Deferred Acceptance (DA) mechanism. PLDA-MQ is fair, satisfies a concept called Priority List-based (PL-) stability, and obtains the student-optimal matching within all PL-stable matchings. Our simulation results show that our new mechanisms are significantly better than simple extensions of the existing mechanisms.

UR - http://www.scopus.com/inward/record.url?scp=85019483203&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85019483203&partnerID=8YFLogxK

U2 - 10.1016/j.artint.2017.04.006

DO - 10.1016/j.artint.2017.04.006

M3 - Article

AN - SCOPUS:85019483203

VL - 249

SP - 47

EP - 71

JO - Artificial Intelligence

JF - Artificial Intelligence

SN - 0004-3702

ER -