schliessen

Filtern

 

Bibliotheken

Bayesian Dynamic Pricing Policies: Learning and Earning Under a Binary Prior Distribution

Motivated by applications in financial services, we consider a seller who offers prices sequentially to a stream of potential customers, observing either success or failure in each sales attempt. The parameters of the underlying demand model are initially unknown, so each price decision involves a t... Full description

Journal Title: Management Science 2012-03, Vol.58 (3), p.570-586
Main Author: HARRISON, J. Michael
Other Authors: BORA KESKIN, N , ZEEVI, Assaf
Format: Electronic Article Electronic Article
Language: English
Subjects:
Publisher: Hanover, MD: INFORMS
ID: ISSN: 0025-1909
Link: http://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=25626038
Zum Text:
SendSend as email Add to Book BagAdd to Book Bag
Staff View
recordid: cdi_gale_businessinsightsgauss_A285090706
title: Bayesian Dynamic Pricing Policies: Learning and Earning Under a Binary Prior Distribution
format: Article
creator:
  • HARRISON, J. Michael
  • BORA KESKIN, N
  • ZEEVI, Assaf
subjects:
  • Analysis
  • Applied sciences
  • Bayesian analysis
  • Bayesian learning
  • Bayesian method
  • Bayesian statistical decision theory
  • Binomial distribution
  • Earnings
  • estimation
  • Exact sciences and technology
  • Experimentation
  • exploitation
  • exploration
  • exploration-exploitation
  • Financial services
  • Firm modelling
  • Incumbents
  • Learning
  • Logical givens
  • Machine learning
  • Management science
  • Marketing
  • Martingales
  • Mathematical models
  • Operational research and scientific management
  • Operational research. Management science
  • Parametric models
  • Portfolio theory
  • Price functions
  • Pricing
  • Pricing policies
  • Revenue
  • Revenue management
  • Statistical methods
  • Studies
  • Trade-off
  • Usage
ispartof: Management Science, 2012-03, Vol.58 (3), p.570-586
description: Motivated by applications in financial services, we consider a seller who offers prices sequentially to a stream of potential customers, observing either success or failure in each sales attempt. The parameters of the underlying demand model are initially unknown, so each price decision involves a trade-off between learning and earning. Attention is restricted to the simplest kind of model uncertainty, where one of two demand models is known to apply, and we focus initially on performance of the myopic Bayesian policy (MBP), variants of which are commonly used in practice. Because learning is passive under the MBP (that is, learning only takes place as a by-product of actions that have a different purpose), it can lead to incomplete learning and poor profit performance. However, under one additional assumption, a constrained variant of the myopic policy is shown to have the following strong theoretical virtue: the expected performance gap relative to a clairvoyant who knows the underlying demand model is bounded by a constant as the number of sales attempts becomes large. This paper was accepted by Gérard P. Cachon, stochastic models and simulation.
language: eng
source:
identifier: ISSN: 0025-1909
fulltext: no_fulltext
issn:
  • 0025-1909
  • 1526-5501
url: Link


@attributes
NO1
SEARCH_ENGINEprimo_central_multiple_fe
SEARCH_ENGINE_TYPEPrimo Central Search Engine
RANK2.6423821
LOCALfalse
PrimoNMBib
record
control
sourceidgale_cross
recordidTN_cdi_gale_businessinsightsgauss_A285090706
sourceformatXML
sourcesystemPC
galeidA285090706
jstor_id41431671
sourcerecordidA285090706
originalsourceidFETCH-LOGICAL-1693t-55906b4b4402ff0f5cab44028f41c81c9c429e47b695a9f5da1fa0cc9a9921840
addsrcrecordideNqFks2P0zAQxSMEEmXhyg0UgZA40DJ2HCfmtl-wSJXYA3vgZE0cu3iV2l07OfS_x95W1bIqQpGSif17T2_sKYrXBBaEts3ntYtqQUj-ZZQ_KWakpnxe10CeFjMAWs-JAPG8eBHjLQA0bcNnxa8z3Opo0ZUXW4drq8rrYJV1q_LaD6nQ8Uu51BhcXkLXl5f7-sb1OpRYnlmHYZtVPpQXNo7BdtNovXtZPDM4RP1q_z0pbr5e_jy_mi9_fPt-frqcEy6qMeUTwDvWMQbUGDC1wvu6NYyoliihGBWaNR0XNQpT90gMglIChaCkZXBSXO18_UY7tEHLTbDrlEl6tLJ3epS-l5zVUvGKY0cqaDn0fcPAGOxopxvTKyEEJKuPO6tN8HeTjqNc26j0MKDTfoqSAOW8YbSqEvruEXrrp-BSo1Jwlrn7aO930AoHLa0zfgyosqc8pW0NAhrgiVocodLT63Qf3mlj0_pfgrdHBPIh8OkB0E3ROh3TK9rV7zGucIrxaAAVfIxBm8MJEpB5tGQeLZlHS-bRSoIP--YxKhxMQKdsPKhozRNUtYljj4yVHTEPRwpsh3_bv9nJbuPow8GWEVYR3pC0P9_t59bDOv4v7h9qK_LW
sourcetypeOpen Access Repository
isCDItrue
recordtypearticle
pqid964026684
display
typearticle
titleBayesian Dynamic Pricing Policies: Learning and Earning Under a Binary Prior Distribution
creatorHARRISON, J. Michael ; BORA KESKIN, N ; ZEEVI, Assaf
creatorcontribHARRISON, J. Michael ; BORA KESKIN, N ; ZEEVI, Assaf
descriptionMotivated by applications in financial services, we consider a seller who offers prices sequentially to a stream of potential customers, observing either success or failure in each sales attempt. The parameters of the underlying demand model are initially unknown, so each price decision involves a trade-off between learning and earning. Attention is restricted to the simplest kind of model uncertainty, where one of two demand models is known to apply, and we focus initially on performance of the myopic Bayesian policy (MBP), variants of which are commonly used in practice. Because learning is passive under the MBP (that is, learning only takes place as a by-product of actions that have a different purpose), it can lead to incomplete learning and poor profit performance. However, under one additional assumption, a constrained variant of the myopic policy is shown to have the following strong theoretical virtue: the expected performance gap relative to a clairvoyant who knows the underlying demand model is bounded by a constant as the number of sales attempts becomes large. This paper was accepted by Gérard P. Cachon, stochastic models and simulation.
identifier
0ISSN: 0025-1909
1EISSN: 1526-5501
2DOI: 10.1287/mnsc.1110.1426
3CODEN: MSCIAM
languageeng
publisherHanover, MD: INFORMS
subjectAnalysis ; Applied sciences ; Bayesian analysis ; Bayesian learning ; Bayesian method ; Bayesian statistical decision theory ; Binomial distribution ; Earnings ; estimation ; Exact sciences and technology ; Experimentation ; exploitation ; exploration ; exploration-exploitation ; Financial services ; Firm modelling ; Incumbents ; Learning ; Logical givens ; Machine learning ; Management science ; Marketing ; Martingales ; Mathematical models ; Operational research and scientific management ; Operational research. Management science ; Parametric models ; Portfolio theory ; Price functions ; Pricing ; Pricing policies ; Revenue ; Revenue management ; Statistical methods ; Studies ; Trade-off ; Usage
ispartofManagement Science, 2012-03, Vol.58 (3), p.570-586
rights
02012 INFORMS
12015 INIST-CNRS
2COPYRIGHT 2012 Institute for Operations Research and the Management Sciences
3Copyright Institute for Operations Research and the Management Sciences Mar 2012
lds50peer_reviewed
oafree_for_read
citedbyFETCH-LOGICAL-1693t-55906b4b4402ff0f5cab44028f41c81c9c429e47b695a9f5da1fa0cc9a9921840
citesFETCH-LOGICAL-1693t-55906b4b4402ff0f5cab44028f41c81c9c429e47b695a9f5da1fa0cc9a9921840
links
openurl$$Topenurl_article
thumbnail$$Usyndetics_thumb_exl
backlink$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=25626038$$DView record in Pascal Francis
search
creatorcontrib
0HARRISON, J. Michael
1BORA KESKIN, N
2ZEEVI, Assaf
title
0Bayesian Dynamic Pricing Policies: Learning and Earning Under a Binary Prior Distribution
1Management Science
descriptionMotivated by applications in financial services, we consider a seller who offers prices sequentially to a stream of potential customers, observing either success or failure in each sales attempt. The parameters of the underlying demand model are initially unknown, so each price decision involves a trade-off between learning and earning. Attention is restricted to the simplest kind of model uncertainty, where one of two demand models is known to apply, and we focus initially on performance of the myopic Bayesian policy (MBP), variants of which are commonly used in practice. Because learning is passive under the MBP (that is, learning only takes place as a by-product of actions that have a different purpose), it can lead to incomplete learning and poor profit performance. However, under one additional assumption, a constrained variant of the myopic policy is shown to have the following strong theoretical virtue: the expected performance gap relative to a clairvoyant who knows the underlying demand model is bounded by a constant as the number of sales attempts becomes large. This paper was accepted by Gérard P. Cachon, stochastic models and simulation.
subject
0Analysis
1Applied sciences
2Bayesian analysis
3Bayesian learning
4Bayesian method
5Bayesian statistical decision theory
6Binomial distribution
7Earnings
8estimation
9Exact sciences and technology
10Experimentation
11exploitation
12exploration
13exploration-exploitation
14Financial services
15Firm modelling
16Incumbents
17Learning
18Logical givens
19Machine learning
20Management science
21Marketing
22Martingales
23Mathematical models
24Operational research and scientific management
25Operational research. Management science
26Parametric models
27Portfolio theory
28Price functions
29Pricing
30Pricing policies
31Revenue
32Revenue management
33Statistical methods
34Studies
35Trade-off
36Usage
issn
00025-1909
11526-5501
fulltextfalse
rsrctypearticle
creationdate2012
recordtypearticle
recordideNqFks2P0zAQxSMEEmXhyg0UgZA40DJ2HCfmtl-wSJXYA3vgZE0cu3iV2l07OfS_x95W1bIqQpGSif17T2_sKYrXBBaEts3ntYtqQUj-ZZQ_KWakpnxe10CeFjMAWs-JAPG8eBHjLQA0bcNnxa8z3Opo0ZUXW4drq8rrYJV1q_LaD6nQ8Uu51BhcXkLXl5f7-sb1OpRYnlmHYZtVPpQXNo7BdtNovXtZPDM4RP1q_z0pbr5e_jy_mi9_fPt-frqcEy6qMeUTwDvWMQbUGDC1wvu6NYyoliihGBWaNR0XNQpT90gMglIChaCkZXBSXO18_UY7tEHLTbDrlEl6tLJ3epS-l5zVUvGKY0cqaDn0fcPAGOxopxvTKyEEJKuPO6tN8HeTjqNc26j0MKDTfoqSAOW8YbSqEvruEXrrp-BSo1Jwlrn7aO930AoHLa0zfgyosqc8pW0NAhrgiVocodLT63Qf3mlj0_pfgrdHBPIh8OkB0E3ROh3TK9rV7zGucIrxaAAVfIxBm8MJEpB5tGQeLZlHS-bRSoIP--YxKhxMQKdsPKhozRNUtYljj4yVHTEPRwpsh3_bv9nJbuPow8GWEVYR3pC0P9_t59bDOv4v7h9qK_LW
startdate201203
enddate201203
creator
0HARRISON, J. Michael
1BORA KESKIN, N
2ZEEVI, Assaf
general
0INFORMS
1Institute for Operations Research and the Management Sciences
scope
0IQODW
1AAYXX
2CITATION
3IOF
48BJ
5FQK
6JBE
7BOBZL
8CLFQK
sort
creationdate201203
titleBayesian Dynamic Pricing Policies: Learning and Earning Under a Binary Prior Distribution
authorHARRISON, J. Michael ; BORA KESKIN, N ; ZEEVI, Assaf
facets
frbrtype5
frbrgroupidcdi_FETCH-LOGICAL-1693t-55906b4b4402ff0f5cab44028f41c81c9c429e47b695a9f5da1fa0cc9a9921840
rsrctypearticles
prefilterarticles
languageeng
creationdate2012
topic
0Analysis
1Applied sciences
2Bayesian analysis
3Bayesian learning
4Bayesian method
5Bayesian statistical decision theory
6Binomial distribution
7Earnings
8estimation
9Exact sciences and technology
10Experimentation
11exploitation
12exploration
13exploration-exploitation
14Financial services
15Firm modelling
16Incumbents
17Learning
18Logical givens
19Machine learning
20Management science
21Marketing
22Martingales
23Mathematical models
24Operational research and scientific management
25Operational research. Management science
26Parametric models
27Portfolio theory
28Price functions
29Pricing
30Pricing policies
31Revenue
32Revenue management
33Statistical methods
34Studies
35Trade-off
36Usage
toplevelpeer_reviewed
creatorcontrib
0HARRISON, J. Michael
1BORA KESKIN, N
2ZEEVI, Assaf
collection
0Pascal-Francis
1CrossRef
2Gale General OneFile
3International Bibliography of the Social Sciences (IBSS)
4International Bibliography of the Social Sciences
5International Bibliography of the Social Sciences
6OpenAIRE (Open Access)
7OpenAIRE
jtitleManagement Science
delivery
delcategoryRemote Search Resource
fulltextno_fulltext
addata
au
0HARRISON, J. Michael
1BORA KESKIN, N
2ZEEVI, Assaf
formatjournal
genrearticle
ristypeJOUR
atitleBayesian Dynamic Pricing Policies: Learning and Earning Under a Binary Prior Distribution
jtitleManagement Science
date2012-03
risdate2012
volume58
issue3
spage570
epage586
pages570-586
issn0025-1909
eissn1526-5501
codenMSCIAM
abstractMotivated by applications in financial services, we consider a seller who offers prices sequentially to a stream of potential customers, observing either success or failure in each sales attempt. The parameters of the underlying demand model are initially unknown, so each price decision involves a trade-off between learning and earning. Attention is restricted to the simplest kind of model uncertainty, where one of two demand models is known to apply, and we focus initially on performance of the myopic Bayesian policy (MBP), variants of which are commonly used in practice. Because learning is passive under the MBP (that is, learning only takes place as a by-product of actions that have a different purpose), it can lead to incomplete learning and poor profit performance. However, under one additional assumption, a constrained variant of the myopic policy is shown to have the following strong theoretical virtue: the expected performance gap relative to a clairvoyant who knows the underlying demand model is bounded by a constant as the number of sales attempts becomes large. This paper was accepted by Gérard P. Cachon, stochastic models and simulation.
copHanover, MD
pubINFORMS
doi10.1287/mnsc.1110.1426
oafree_for_read