{"id":{"repo_id":"bu","oai_identifier":"oai:open.bu.edu:2144/53304"},"canonical_url":"https://search.ndltd.org/etd/bu/oai:open.bu.edu:2144/53304","repository":{"repo_id":"bu","name":"Boston University","base_url":"https://open.bu.edu/oai/request"},"display":{"title":"Essays on experimentation, dynamic advice, and shill bidding","abstract":"Chapter 1: We study a dynamic delegation problem where a long-lived principal delegates experimentation to a sequence of short-lived agents. Each agent is active for one period before being replaced. Because the principal values future outcomes more, she has stronger incentives to experiment than the agents. We model this as a contextual bandit problem, where experimentation outcomes follow a type-dependent Poisson process. Each agent’s type affects the success probability: one type has known success probability, while the other is uncertain and could be either high or low. Players gradually learn through experimentation. Focusing on stopping-time policies, we show that experimentation follows cyclical dynamics, alternating between engagement, abandonment, and revival. Misaligned incentives between the principal and agents lead to both over- and under-experimentation. Chapter 2: This paper develops a dynamic model of expert advice in markets for credence and experience goods. Each period, a seller (the expert) sets prices for two products and recommends one to a consumer. Consumers rely on expert advice because they do not know which product fits best until after purchase. However, the expert's ability -- whether skilled or mediocre -- is private information. Skilled experts receive accurate signals about product fit, while mediocre experts have no useful information. This creates a credence good problem, where consumers cannot fully verify whether the recommendation was beneficial. I characterize the equilibria of this advice game and identify conditions under which a separating equilibrium is expert-optimal. Chapter 3: This paper examines whether shill bidding can increase revenue in a common value button auction, where dropout prices and bidder identities are unobservable. We characterize the symmetric equilibrium under two cases: when the seller uses a secret reserve price, and when the seller acts as a shill bidder. Shill bidding affects revenue by altering effective participation, which distorts bidders’ expected value of the item. Whether shill bidding or a secret reserve price is more profitable depends on the number of bidders.","abstract_html":"Chapter 1: We study a dynamic delegation problem where a long-lived principal delegates experimentation to a sequence of short-lived agents. Each agent is active for one period before being replaced. Because the principal values future outcomes more, she has stronger incentives to experiment than the agents. We model this as a contextual bandit problem, where experimentation outcomes follow a type-dependent Poisson process. Each agent’s type affects the success probability: one type has known success probability, while the other is uncertain and could be either high or low. Players gradually learn through experimentation. Focusing on stopping-time policies, we show that experimentation follows cyclical dynamics, alternating between engagement, abandonment, and revival. Misaligned incentives between the principal and agents lead to both over- and under-experimentation. Chapter 2: This paper develops a dynamic model of expert advice in markets for credence and experience goods. Each period, a seller (the expert) sets prices for two products and recommends one to a consumer. Consumers rely on expert advice because they do not know which product fits best until after purchase. However, the expert&#x27;s ability -- whether skilled or mediocre -- is private information. Skilled experts receive accurate signals about product fit, while mediocre experts have no useful information. This creates a credence good problem, where consumers cannot fully verify whether the recommendation was beneficial. I characterize the equilibria of this advice game and identify conditions under which a separating equilibrium is expert-optimal. Chapter 3: This paper examines whether shill bidding can increase revenue in a common value button auction, where dropout prices and bidder identities are unobservable. We characterize the symmetric equilibrium under two cases: when the seller uses a secret reserve price, and when the seller acts as a shill bidder. Shill bidding affects revenue by altering effective participation, which distorts bidders’ expected value of the item. Whether shill bidding or a secret reserve price is more profitable depends on the number of bidders.","abstract_has_math":false,"creators":["Liu, Zixian"],"institution":"Boston University","degree_name":"Doctor of Philosophy","degree_level":"doctoral","degree_discipline":"Economics","degree_department":null,"school":null,"contributors":[],"advisors":["Ortner, Juan","Ma, Albert"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025","date_published":"2025","updated_at":"2026-11-18T19:38:11Z","subjects":["Economic theory"],"languages":["en_US"],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2144/53304","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"https://hdl.handle.net/2144/53304","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://open.bu.edu/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Aopen.bu.edu%3A2144%2F53304","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Ortner, Juan","Ma, Albert"]},{"key":"dc:creator","label":"Author","values":["Liu, Zixian"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-07-13T15:29:58Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2026-07-13T15:29:58Z"]},{"key":"dc:date.issued","label":"Date","values":["2025"]},{"key":"dc:date.updated","label":"Dc Date Updated","values":["2026-07-10T19:15:26Z"]},{"key":"dc:type","label":"Dc Type","values":["Thesis/Dissertation"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Economics"]},{"key":"thesis:degree_level","label":"Degree Level","values":["doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Boston University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Economic theory"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en_US"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/2144/53304"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["2025"]},{"key":"dc:description.abstract","label":"Abstract","values":["Chapter 1: We study a dynamic delegation problem where a long-lived principal delegates experimentation to a sequence of short-lived agents. Each agent is active for one period before being replaced. Because the principal values future outcomes more, she has stronger incentives to experiment than the agents. We model this as a contextual bandit problem, where experimentation outcomes follow a type-dependent Poisson process. Each agent’s type affects the success probability: one type has known success probability, while the other is uncertain and could be either high or low. Players gradually learn through experimentation. Focusing on stopping-time policies, we show that experimentation follows cyclical dynamics, alternating between engagement, abandonment, and revival. Misaligned incentives between the principal and agents lead to both over- and under-experimentation. Chapter 2: This paper develops a dynamic model of expert advice in markets for credence and experience goods. Each period, a seller (the expert) sets prices for two products and recommends one to a consumer. Consumers rely on expert advice because they do not know which product fits best until after purchase. However, the expert's ability -- whether skilled or mediocre -- is private information. Skilled experts receive accurate signals about product fit, while mediocre experts have no useful information. This creates a credence good problem, where consumers cannot fully verify whether the recommendation was beneficial. I characterize the equilibria of this advice game and identify conditions under which a separating equilibrium is expert-optimal. Chapter 3: This paper examines whether shill bidding can increase revenue in a common value button auction, where dropout prices and bidder identities are unobservable. We characterize the symmetric equilibrium under two cases: when the seller uses a secret reserve price, and when the seller acts as a shill bidder. Shill bidding affects revenue by altering effective participation, which distorts bidders’ expected value of the item. Whether shill bidding or a secret reserve price is more profitable depends on the number of bidders."]},{"key":"dc:title","label":"Title","values":["Essays on experimentation, dynamic advice, and shill bidding"]}]}],"canonical_facts":{"dc:contributor.advisor":["Ortner, Juan","Ma, Albert"],"dc:creator":["Liu, Zixian"],"dc:date.accessioned":["2026-07-13T15:29:58Z"],"dc:date.available":["2026-07-13T15:29:58Z"],"dc:date.issued":["2025"],"dc:date.updated":["2026-07-10T19:15:26Z"],"dc:description":["2025"],"dc:description.abstract":["Chapter 1: We study a dynamic delegation problem where a long-lived principal delegates experimentation to a sequence of short-lived agents. Each agent is active for one period before being replaced. Because the principal values future outcomes more, she has stronger incentives to experiment than the agents. We model this as a contextual bandit problem, where experimentation outcomes follow a type-dependent Poisson process. Each agent’s type affects the success probability: one type has known success probability, while the other is uncertain and could be either high or low. Players gradually learn through experimentation. Focusing on stopping-time policies, we show that experimentation follows cyclical dynamics, alternating between engagement, abandonment, and revival. Misaligned incentives between the principal and agents lead to both over- and under-experimentation. Chapter 2: This paper develops a dynamic model of expert advice in markets for credence and experience goods. Each period, a seller (the expert) sets prices for two products and recommends one to a consumer. Consumers rely on expert advice because they do not know which product fits best until after purchase. However, the expert's ability -- whether skilled or mediocre -- is private information. Skilled experts receive accurate signals about product fit, while mediocre experts have no useful information. This creates a credence good problem, where consumers cannot fully verify whether the recommendation was beneficial. I characterize the equilibria of this advice game and identify conditions under which a separating equilibrium is expert-optimal. Chapter 3: This paper examines whether shill bidding can increase revenue in a common value button auction, where dropout prices and bidder identities are unobservable. We characterize the symmetric equilibrium under two cases: when the seller uses a secret reserve price, and when the seller acts as a shill bidder. Shill bidding affects revenue by altering effective participation, which distorts bidders’ expected value of the item. Whether shill bidding or a secret reserve price is more profitable depends on the number of bidders."],"dc:identifier.uri":["https://hdl.handle.net/2144/53304"],"dc:language.iso":["en_US"],"dc:subject":["Economic theory"],"dc:title":["Essays on experimentation, dynamic advice, and shill bidding"],"dc:type":["Thesis/Dissertation"],"thesis:degree_discipline":["Economics"],"thesis:degree_level":["doctoral"],"thesis:degree_name":["Doctor of Philosophy"],"thesis:institution_name":["Boston University"]},"updated_at":"2026-11-18T19:38:11Z"}