본문

서브메뉴

Essays on the Economics of Search Algorithms
Essays on the Economics of Search Algorithms
Essays on the Economics of Search Algorithms

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20250211151316
ISBN  
9798382742991
DDC  
330.9
저자명  
Monk, Kyle.
서명/저자  
Essays on the Economics of Search Algorithms
발행사항  
[Sl] : Georgetown University, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
159 p
주기사항  
Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
주기사항  
Advisor: Rust, John.
학위논문주기  
Thesis (Ph.D.)--Georgetown University, 2024.
초록/해제  
요약This dissertation explores the economics of search algorithms deployed by platforms. Chapters 1 and 2 address challenges in evaluating market power associated with algorithms. Chapter 3 explores the mechanisms that platforms use to alter search outcomes when their access to user data changes.In Chapter 1, I develop a model of search based on competition over algorithms that can be used to evaluate the market power of an algorithm that aids a consumer's search process. In the model, platforms make an endogenous choice about how much their algorithms should favor consumers' preferences versus the platform's profitability-per-consumer. This decision is modeled as a choice to set a distance between realized, consumer-optimal, and platform-optimal search outcomes. Finally, a general framework to apply this model across platform types is presented. In Chapter 2, the model is applied to JD.com e-commerce data. First, I estimate a discrete choice model of consumer purchase decisions given their search outcomes. Then, I estimate the structural search model. I find that the algorithms achieve an outcome almost identical to the outcomes that maximize the platform's profitability-per-user. The finding suggests one of two possibilities: JD.com has tremendous market power in the market studied, or JD.com is using a suboptimal ranking algorithm that maximizes profits-per-consumer rather than total profits. Realigning the incentives of the platform and consumers by banning first-party participation on e-commerce platforms would improve the market outcomes for consumers and sellers, however, it would decrease the platform's profitability-per-consumer by a larger magnitude. In Chapter 3, I calibrate a structural model of consumer search that allows the platform's beliefs about the consumer to be incorrect. The model estimates the weights the platform places on user preferences and expected profits when choosing to display search results. I estimate this model separately for JD.com shoppers that the platform possesses predictive data about and for those without this data. Then, I simulate search and purchase outcomes for the shoppers with predictive data as if this data did not exist. Finally, I decompose the effects into those from price discrimination, those from personalized steering, and those from changes in market power.
키워드  
Algorithms
키워드  
Digital economics
키워드  
Industrial organization
키워드  
Market power
키워드  
Personalized pricing
기타저자  
Georgetown University Economics
기본자료저록  
Dissertations Abstracts International. 85-11B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017161147
■00520250211151316
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798382742991
■035    ▼a(MiAaPQ)AAI31238506
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a330.9
■1001  ▼aMonk,  Kyle.
■24510▼aEssays  on  the  Economics  of  Search  Algorithms
■260    ▼a[Sl]▼bGeorgetown  University▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a159  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  85-11,  Section:  B.
■500    ▼aAdvisor:  Rust,  John.
■5021  ▼aThesis  (Ph.D.)--Georgetown  University,  2024.
■520    ▼aThis  dissertation  explores  the  economics  of  search  algorithms  deployed  by  platforms.  Chapters  1  and  2  address  challenges  in  evaluating  market  power  associated  with  algorithms.  Chapter  3  explores  the  mechanisms  that  platforms  use  to  alter  search  outcomes  when  their  access  to  user  data  changes.In  Chapter  1,  I  develop  a  model  of  search  based  on  competition  over  algorithms  that  can  be  used  to  evaluate  the  market  power  of  an  algorithm  that  aids  a  consumer's  search  process.  In  the  model,  platforms  make  an  endogenous  choice  about  how  much  their  algorithms  should  favor  consumers'  preferences  versus  the  platform's  profitability-per-consumer.  This  decision  is  modeled  as  a  choice  to  set  a  distance  between  realized,  consumer-optimal,  and  platform-optimal  search  outcomes.  Finally,  a  general  framework  to  apply  this  model  across  platform  types  is  presented.  In  Chapter  2,  the  model  is  applied  to  JD.com  e-commerce  data.  First,  I  estimate  a  discrete  choice  model  of  consumer  purchase  decisions  given  their  search  outcomes.  Then,  I  estimate  the  structural  search  model.  I  find  that  the  algorithms  achieve  an  outcome  almost  identical  to  the  outcomes  that  maximize  the  platform's  profitability-per-user.  The  finding  suggests  one  of  two  possibilities:  JD.com  has  tremendous  market  power  in  the  market  studied,  or  JD.com  is  using  a  suboptimal  ranking  algorithm  that  maximizes  profits-per-consumer  rather  than  total  profits.  Realigning  the  incentives  of  the  platform  and  consumers  by  banning  first-party  participation  on  e-commerce  platforms  would  improve  the  market  outcomes  for  consumers  and  sellers,  however,  it  would  decrease  the  platform's  profitability-per-consumer  by  a  larger  magnitude.  In  Chapter  3,  I  calibrate  a  structural  model  of  consumer  search  that  allows  the  platform's  beliefs  about  the  consumer  to  be  incorrect.  The  model  estimates  the  weights  the  platform  places  on  user  preferences  and  expected  profits  when  choosing  to  display  search  results.  I  estimate  this  model  separately  for  JD.com  shoppers  that  the  platform  possesses  predictive  data  about  and  for  those  without  this  data.  Then,  I  simulate  search  and  purchase  outcomes  for  the  shoppers  with  predictive  data  as  if  this  data  did  not  exist.  Finally,  I  decompose  the  effects  into  those  from  price  discrimination,  those  from  personalized  steering,  and  those  from  changes  in  market  power.
■590    ▼aSchool  code:  0076.
■653    ▼aAlgorithms
■653    ▼aDigital  economics
■653    ▼aIndustrial  organization
■653    ▼aMarket  power
■653    ▼aPersonalized  pricing
■690    ▼a0501
■690    ▼a0796
■690    ▼a0338
■71020▼aGeorgetown  University▼bEconomics.
■7730  ▼tDissertations  Abstracts  International▼g85-11B.
■790    ▼a0076
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17161147▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

미리보기

내보내기

chatGPT토론

Ai 추천 관련 도서


    신착도서 더보기
    최근 3년간 통계입니다.

    소장정보

    • 예약
    • 소재불명신고
    • 나의폴더
    • 우선정리요청
    • 비도서대출신청
    • 야간 도서대출신청
    소장자료
    등록번호 청구기호 소장처 대출가능여부 대출정보
    TF11924 전자도서 대출가능 마이폴더 부재도서신고 비도서대출신청 야간 도서대출신청

    * 대출중인 자료에 한하여 예약이 가능합니다. 예약을 원하시면 예약버튼을 클릭하십시오.

    해당 도서를 다른 이용자가 함께 대출한 도서

    관련 인기도서

    로그인 후 이용 가능합니다.