서브메뉴
검색
Essays on Mechanism Design
Essays on Mechanism Design
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202103520
- ISBN
- 9798315798361
- DDC
- 658
- 저자명
- Liu, Yijun.
- 서명/저자
- Essays on Mechanism Design
- 발행사항
- [Sl] : Northwestern University, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 175 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-12, Section: A.
- 주기사항
- Advisor: Ely, Jeffrey;Pavan, Alessandro.
- 학위논문주기
- Thesis (Ph.D.)--Northwestern University, 2025.
- 초록/해제
- 요약This dissertation comprises three essays on mechanism design. The first essay investigates a dynamic mechanism design problem in which the agent possesses evolving private information and is subject to limited liability. The second essay examines a dynamic delegation problem in which a principal allocates resources to a biased agent over two periods and receives an interim signal about the agent's type. The third essay analyzes an optimal market segmentation problem under a monotone partition constraint. This framework is applied to university grading and offers a potential explanation for the phenomenon of GPA inflation.Chapter 1 studies a dynamic screening model where a principal hires an agent with limited liability. The agent's private cost of working is an i.i.d. draw from a continuous distribution. His working status is publicly observable. The limited liability constraint requires that payments remain nonnegative at all times. In this setting, despite costs being i.i.d. and the payoffs being additively separable across periods, the optimal mechanism does not treat each period independently. Instead, it features backloading payments and requires the agent to work in consecutive periods, which provides an explanation for many real-world practices of offering rewards based on the number of consecutive actions, such as consecutive rides, purchases, or logins in ride-sharing, retail, and gaming industries. Specifically, I characterize conditions under which the optimal mechanism either grants the agent flexibility to start working in any period or restricts the starting period to the first. In either case, once the agent begins working, he is incentivized to work consecutively until the end.In Chapter 2, there is a principal who allocates resources to an agent with a project of privately known quality. The principal aims at allocating resources to match the project's quality, while the agent always prefers a higher allocation. In the presence of a midterm review, Chapter 2 characterizes the principal's optimal resource allocation strategy over two periods (before and after the review). Given the time lag between the midterm review and the initial decision-making, the principal should leverage the midterm review information to inform the first-period allocation. Specifically, the optimal first-period allocation aligns with the project's reported quality, with higher-quality projects receiving more resources. In the second period, the principal adjusts the allocation to reward or penalize the agent based on the midterm review-being more lenient toward lower types and stricter with higher types. This strategy incentives the agent to report truthfully in the first period and maximizes the principal's expected payoff. The benefits of this levering-information strategy are robust across a wide class of loss functions.Based on [13], Chapter 3 studies optimal market segmentation under the constraint that only monotone partitions of the type space are allowed. Specifically, consider a setting in which a university observes students' private types and designs a grading rule to disclose information to employers. For fairness, the grading rule must be a monotone partition, meaning higher types receive higher grades. When the university's objective is a weighted sum of students' and employer's surplus, Chapter 3 finds that the optimal grading rule either fully reveals all type information or features coarser partitions for higher types and finer partitions for lower types. The latter structure is optimal when the university places more weight on students' surplus, providing a theoretical explanation for the GPA inflation observed in practice.
- 일반주제명
- Finance
- 키워드
- Contract theory
- 키워드
- Mechanism design
- 기타저자
- Northwestern University Economics
- 기본자료저록
- Dissertations Abstracts International. 86-12A.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017357492
■00520260202103520
■006m o d
■007cr#unu||||||||
■020 ▼a9798315798361
■035 ▼a(MiAaPQ)AAI32038566
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a658
■1001 ▼aLiu, Yijun.▼0(orcid)0009-0001-6981-7508
■24510▼aEssays on Mechanism Design
■260 ▼a[Sl]▼bNorthwestern University▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a175 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-12, Section: A.
■500 ▼aAdvisor: Ely, Jeffrey;Pavan, Alessandro.
■5021 ▼aThesis (Ph.D.)--Northwestern University, 2025.
■520 ▼aThis dissertation comprises three essays on mechanism design. The first essay investigates a dynamic mechanism design problem in which the agent possesses evolving private information and is subject to limited liability. The second essay examines a dynamic delegation problem in which a principal allocates resources to a biased agent over two periods and receives an interim signal about the agent's type. The third essay analyzes an optimal market segmentation problem under a monotone partition constraint. This framework is applied to university grading and offers a potential explanation for the phenomenon of GPA inflation.Chapter 1 studies a dynamic screening model where a principal hires an agent with limited liability. The agent's private cost of working is an i.i.d. draw from a continuous distribution. His working status is publicly observable. The limited liability constraint requires that payments remain nonnegative at all times. In this setting, despite costs being i.i.d. and the payoffs being additively separable across periods, the optimal mechanism does not treat each period independently. Instead, it features backloading payments and requires the agent to work in consecutive periods, which provides an explanation for many real-world practices of offering rewards based on the number of consecutive actions, such as consecutive rides, purchases, or logins in ride-sharing, retail, and gaming industries. Specifically, I characterize conditions under which the optimal mechanism either grants the agent flexibility to start working in any period or restricts the starting period to the first. In either case, once the agent begins working, he is incentivized to work consecutively until the end.In Chapter 2, there is a principal who allocates resources to an agent with a project of privately known quality. The principal aims at allocating resources to match the project's quality, while the agent always prefers a higher allocation. In the presence of a midterm review, Chapter 2 characterizes the principal's optimal resource allocation strategy over two periods (before and after the review). Given the time lag between the midterm review and the initial decision-making, the principal should leverage the midterm review information to inform the first-period allocation. Specifically, the optimal first-period allocation aligns with the project's reported quality, with higher-quality projects receiving more resources. In the second period, the principal adjusts the allocation to reward or penalize the agent based on the midterm review-being more lenient toward lower types and stricter with higher types. This strategy incentives the agent to report truthfully in the first period and maximizes the principal's expected payoff. The benefits of this levering-information strategy are robust across a wide class of loss functions.Based on [13], Chapter 3 studies optimal market segmentation under the constraint that only monotone partitions of the type space are allowed. Specifically, consider a setting in which a university observes students' private types and designs a grading rule to disclose information to employers. For fairness, the grading rule must be a monotone partition, meaning higher types receive higher grades. When the university's objective is a weighted sum of students' and employer's surplus, Chapter 3 finds that the optimal grading rule either fully reveals all type information or features coarser partitions for higher types and finer partitions for lower types. The latter structure is optimal when the university places more weight on students' surplus, providing a theoretical explanation for the GPA inflation observed in practice.
■590 ▼aSchool code: 0163.
■650 4▼aFinance
■653 ▼aContract theory
■653 ▼aDynamic delegation
■653 ▼aDynamic mechanism design
■653 ▼aInformation economics
■653 ▼aMechanism design
■690 ▼a0501
■690 ▼a0511
■690 ▼a0508
■71020▼aNorthwestern University▼bEconomics.
■7730 ▼tDissertations Abstracts International▼g86-12A.
■790 ▼a0163
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17357492▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


