본문

서브메뉴

Essays on Pricing and Algorithmic Decision-Making
Essays on Pricing and Algorithmic Decision-Making
Essays on Pricing and Algorithmic Decision-Making

Detailed Information

자료유형  
 학위논문 서양
최종처리일시  
20260202103058
ISBN  
9798280756755
DDC  
004
저자명  
Zhao, Hangcheng.
서명/저자  
Essays on Pricing and Algorithmic Decision-Making
발행사항  
[Sl] : University of Pennsylvania, 2025
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2025
형태사항  
182 p
주기사항  
Source: Dissertations Abstracts International, Volume: 86-12, Section: B.
주기사항  
Advisor: Berman, Ron.
학위논문주기  
Thesis (Ph.D.)--University of Pennsylvania, 2025.
초록/해제  
요약This dissertation investigates three different aspects of pricing, employing different methodologies (algorithmic decision-making using reinforcement learning and recommendation systems using information design) and in various substantive areas (an e-commerce digital platform setting, online advertising, or the healthcare market).Chapter 1 empirically investigates the impact of firms using learning algorithms for pricing and advertising decisions in an e-commerce digital platform setting. Previous research suggests that when competing firms use algorithms for pricing decisions, the competition often results in tacit collusion-the algorithms learn to settle on higher than competitive prices, which increase firm profits but hurt consumers. However, in real business scenarios, firms need to make their pricing and advertising decisions together, and the outcomes of such two-dimensional decision-making in a competitive environment are unclear ex-ante without simulating the outcomes experimentally. Therefore, we conduct extensive multi-agent reinforcement learning simulation experiments, calibrated to estimates from large-scale keyword search data from Amazon.com. The results show that algorithms can facilitate "beneficial collusion," resulting in "win-win-win" outcomes for consumers, sellers, and the platform. We collected and analyzed a large-scale high-frequency product keyword search dataset fromAmazon.com, which includes more than two thousand highly searched keywords. The dataset contains more than two million products and is representative of search results containing both sponsored and organic products from all IP addresses in the US. We estimate consumer consideration size, generate an algorithm usage index based on the correlation patterns in prices and find a negative interaction between the estimated consumer search costs and the algorithm usage index, providing empirical evidence of beneficial collusion.Chapter 2 examines pricing in the US healthcare market by empirically analyzing a large-scale nationwide proprietary healthcare insurance claims dataset. We ask how increased price transparency influences consumer shopping behavior and, ultimately, market prices. We find that potential savings from price shopping are expected to be limited after taking into consideration a few realistic factors in the market. This study presents an important setting where reducing information frictions will not necessarily increase consumer price shopping behavior and will not create downward pressure on high prices in the market.Chapter 3 shifts the focus to recommendation algorithms that firms can design to strategically provide information to consumers about products with uncertain matches to their tastes. We investigate scenarios where firms can also adjust their pricing strategies and examine whether algorithms that always recommend matching products are beneficial for the firm. We surprisingly find that firms are sometimes better off demarketing (i.e., not recommending a matching product) their products, and this is the case also for platforms.
일반주제명  
Information technology
일반주제명  
Finance
키워드  
Reinforcement learning
키워드  
Algorithm usage index
키워드  
Consumer shopping
키워드  
Decision-making
기타저자  
University of Pennsylvania Marketing
기본자료저록  
Dissertations Abstracts International. 86-12B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008260126s2025        us                              c    eng  d
■001000017356903
■00520260202103058
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798280756755
■035    ▼a(MiAaPQ)AAI31933602
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a004
■1001  ▼aZhao,  Hangcheng.
■24510▼aEssays  on  Pricing  and  Algorithmic  Decision-Making
■260    ▼a[Sl]▼bUniversity  of  Pennsylvania▼c2025
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2025
■300    ▼a182  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  86-12,  Section:  B.
■500    ▼aAdvisor:  Berman,  Ron.
■5021  ▼aThesis  (Ph.D.)--University  of  Pennsylvania,  2025.
■520    ▼aThis  dissertation  investigates  three  different  aspects  of  pricing,  employing  different  methodologies  (algorithmic  decision-making  using  reinforcement  learning  and  recommendation  systems  using  information  design)  and  in  various  substantive  areas  (an  e-commerce  digital  platform  setting,  online  advertising,  or  the  healthcare  market).Chapter  1  empirically  investigates  the  impact  of  firms  using  learning  algorithms  for  pricing  and  advertising  decisions  in  an  e-commerce  digital  platform  setting.  Previous  research  suggests  that  when  competing  firms  use  algorithms  for  pricing  decisions,  the  competition  often  results  in  tacit  collusion-the  algorithms  learn  to  settle  on  higher  than  competitive  prices,  which  increase  firm  profits  but  hurt  consumers.  However,  in  real  business  scenarios,  firms  need  to  make  their  pricing  and  advertising  decisions  together,  and  the  outcomes  of  such  two-dimensional  decision-making  in  a  competitive  environment  are  unclear  ex-ante  without  simulating  the  outcomes  experimentally.  Therefore,  we  conduct  extensive  multi-agent  reinforcement  learning  simulation  experiments,  calibrated  to  estimates  from  large-scale  keyword  search  data  from  Amazon.com.  The  results  show  that  algorithms  can  facilitate  "beneficial  collusion,"  resulting  in  "win-win-win"  outcomes  for  consumers,  sellers,  and  the  platform.  We  collected  and  analyzed  a  large-scale  high-frequency  product  keyword  search  dataset  fromAmazon.com,  which  includes  more  than  two  thousand  highly  searched  keywords.  The  dataset  contains  more  than  two  million  products  and  is  representative  of  search  results  containing  both  sponsored  and  organic  products  from  all  IP  addresses  in  the  US.  We  estimate  consumer  consideration  size,  generate  an  algorithm  usage  index  based  on  the  correlation  patterns  in  prices  and  find  a  negative  interaction  between  the  estimated  consumer  search  costs  and  the  algorithm  usage  index,  providing  empirical  evidence  of  beneficial  collusion.Chapter  2  examines  pricing  in  the  US  healthcare  market  by  empirically  analyzing  a  large-scale  nationwide  proprietary  healthcare  insurance  claims  dataset.  We  ask  how  increased  price  transparency  influences  consumer  shopping  behavior  and,  ultimately,  market  prices.  We  find  that  potential  savings  from  price  shopping  are  expected  to  be  limited  after  taking  into  consideration  a  few  realistic  factors  in  the  market.  This  study  presents  an  important  setting  where  reducing  information  frictions  will  not  necessarily  increase  consumer  price  shopping  behavior  and  will  not  create  downward  pressure  on  high  prices  in  the  market.Chapter  3  shifts  the  focus  to  recommendation  algorithms  that  firms  can  design  to  strategically  provide  information  to  consumers  about  products  with  uncertain  matches  to  their  tastes.  We  investigate  scenarios  where  firms  can  also  adjust  their  pricing  strategies  and  examine  whether  algorithms  that  always  recommend  matching  products  are  beneficial  for  the  firm.  We  surprisingly  find  that  firms  are  sometimes  better  off  demarketing  (i.e.,  not  recommending  a  matching  product)  their  products,  and  this  is  the  case  also  for  platforms.
■590    ▼aSchool  code:  0175.
■650  4▼aInformation  technology
■650  4▼aFinance
■653    ▼aReinforcement  learning
■653    ▼aAlgorithm  usage  index
■653    ▼aConsumer  shopping
■653    ▼aDecision-making
■690    ▼a0338
■690    ▼a0489
■690    ▼a0511
■690    ▼a0800
■690    ▼a0508
■71020▼aUniversity  of  Pennsylvania▼bMarketing.
■7730  ▼tDissertations  Abstracts  International▼g86-12B.
■790    ▼a0175
■791    ▼aPh.D.
■792    ▼a2025
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17356903▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

Preview

Export

ChatGPT Discussion

AI Recommended Related Books


    New Books MORE
    Statistics for the past 3 years. Go to brief

    Buch Status

    • Reservierung
    • frei buchen
    • Meine Mappe
    • Erste Aufräumarbeiten Anfrage
    • Non-Book Loan Application
    • Nighttime Book Loan Application
    Sammlungen
    Registrierungsnummer callnumber Standort Verkehr Status Verkehr Info
    TF19034 전자도서 대출가능 My Folder 부재도서신고 비도서대출신청 야간 도서대출신청

    * Kredite nur für Ihre Daten gebucht werden. Wenn Sie buchen möchten Reservierungen, klicken Sie auf den Button.

    Books borrowed together with this book

    Related Popular Books

    Available after logging in.