1、毕业论文(设计)外文翻译题目网络团购的发展现状与问题探究一、外文原文标题SEGMENTINGUNCERTAINDEMANDINGROUPBUYINGAUCTIONS原文DEMANDUNCERTAINTYISAKEYFACTORINASELLERSDECISIONMAKINGPROCESSFORPRODUCTSSOLDTHROUGHONLINEAUCTIONSWEEXPLOREDEMANDUNCERTAINTYINGROUPBUYINGAUCTIONSINTERMSOFTHEEXTENTOFLOWVALUATIONDEMANDANDHIGHVALUATIONDEMANDWEFOCUSONTHEA
2、NALYSISOFAMONOPOLISTICGROUPBUYINGRETAILERTHATSELLSPRODUCTSTOCONSUMERSWHOEXPRESSDIFFERENTPRODUCTVALUATIONSWEALSOEXAMINETHEPERFORMANCEOFAGROUPBUYINGSELLERWHOFACESCOMPETITIVEPOSTEDPRICESELLERSINAMARKETFORTHESALEOFTHESAMEPRODUCTS,UNDERSIMILARASSUMPTIONSABOUTUNCERTAINDEMANDBASEDONANASHEQUILIBRIUMANALYSIS
3、OFBIDDERSTRATEGIESFORBOTHOFTHESESELLERSIDECOMPETITIONSTRUCTURES,WEAREABLETOCHARACTERIZETHEGROUPBUYINGAUCTIONBIDDERSDOMINANTSTRATEGIESWEOBTAINEDANUMBEROFINTERESTINGFINDINGSGROUPBUYINGISLIKELYTOBEMOREEFFECTIVEINSETTINGSWHERETHEREISLARGERLOWVALUATIONDEMANDTHANHIGHVALUATIONDEMANDTHESTRUCTUREOFDEMANDMATT
4、ERSTHISFINDINGHASRELEVANCETOTHEMARKETPLACEFORNEWCAMERAS,NEXTGENERATIONMICROPROCESSORSANDCOMPUTERS,ANDOTHERHIGHVALUATIONGOODS,WHICHAREUNLIKELYTOBEASEFFECTIVELYSOLDINGROUPBUYINGMARKETSWEOBTAINEDADDITIONALRESULTSFORTHECASEOFCONTINUOUSDEMAND,ANDFINDTHATTHEREISABASISFORTHESELLERTOIMPROVEREVENUESVIAEFFECT
5、IVEGROUPBUYINGAUCTIONPRICECURVEDESIGNKEYWORDSCONSUMERBEHAVIOR,BIDDINGSTRATEGY,DEMANDUNCERTAINTY,ECONOMICANALYSIS,ELECTRONICMARKETS,GROUPBUYINGAUCTIONS,MARKETMECHANISM,POSTEDPRICEMECHANISM,SIMULATION,UNCERTAINTYRISKTHEDEVELOPMENTOFADVANCEDITMAKESITPOSSIBLETOUSENOVELBUSINESSMODELSTOHANDLEBUSINESSPROBL
6、EMSINNEWANDINNOVATIVEWAYSWITHTHEGROWTHOFTHEINTERNET,ANUMBEROFNEWELECTRONICAUCTIONMECHANISMSHAVEEMERGED,ANDAUCTIONSAREGENERALLYKNOWNTOCREATEHIGHEREXPECTEDSELLERREVENUETHANPOSTEDPRICESWHENTHECOSTOFRUNNINGANAUCTIONISMINIMALORCOSTLESSWANG1993SOMEOFTHENEWMECHANISMSWEHAVESEENINCLUDETHEONLINEYANKEEANDDUTCH
7、AUCTIONS,ANDTHE“NAMEYOUROWNPRICE”AND“BUYITNOW”MECHANISMSANEXAMPLEISEBAYSDUTCHAUCTIONFORTHESALEOFMULTIPLEITEMSOFTHESAMEDESCRIPTIONANOTHEROFTHESENEWELECTRONICMARKETMECHANISMSTHATWEHAVEOBSERVEDISTHEGROUPBUYINGAUCTION,AHOMOGENEOUSMULTIUNITAUCTIONMITCHELL2002,LIETAL2004INTERNETBASEDSELLERSANDDIGITALINTER
8、MEDIARIESHAVEADOPTEDTHISMARKETMECHANISMONSITESSUCHASLETSBUYITCOMWWWLETSBUYITCOMANDEWINWINCOMWWWEWINWINCOMTHESESITESOFFERTRANSACTIONMAKINGMECHANISMSTHATAREDIFFERENTFROMTRADITIONALAUCTIONSINTRADITIONALAUCTIONS,BIDDERSCOMPETEAGAINSTONEANOTHERTOBETHEWINNERINGROUPBUYINGAUCTIONS,HOWEVER,BIDDERSHAVEANINCEN
9、TIVETOAGGREGATETHEIRBIDSSOTHATTHESELLERORDIGITALINTERMEDIARYOFFERSALOWERPRICEATWHICHTHEYALLCANBUYTHEDESIREDGOODSHORNETAL2000MCCABEETAL1991HAVEEXPLOREDMULTIUNITVICKREYAUCTIONSINEXPERIMENTALRESEARCH,HOWEVER,THEYDIDNOTCONSIDERTHEPOSSIBILITYOFSTOCHASTICBIDDERARRIVALORDEMANDUNCERTAINTYTHISPAPERISTHEFIRST
10、TOEXAMINETHEIMPACTSOFDEMANDUNCERTAINTYONTHEPERFORMANCEONONLINEGROUPBUYINGAUCTIONSBASEDONANASHEQUILIBRIUMANALYSISOFBIDDERSTRATEGIESFORAMONOPOLISTSELLERANDACOMPETITIVESELLER,WEAREABLETOCHARACTERIZETHEGROUPBUYINGAUCTIONBIDDERSSYMMETRICANDDOMINANTSTRATEGIESWEFINDTHATGROUPBUYINGISLIKELYTOBEMOREEFFECTIVEI
11、NSETTINGSWHERETHEREISLARGERLOWVALUATIONDEMANDTHANHIGHVALUATIONDEMANDTHUS,THESTRUCTUREOFDEMANDATDIFFERENTLEVELOFWILLINGNESSTOPAYBYCONSUMERSMATTERSTHISHASRELEVANCETOTHEMARKETPLACEFORNEWCAMERAS,NEXTGENERATIONMICROPROCESSORSANDCOMPUTERS,ANDOTHERHIGHVALUATIONGOODSWEOBTAINEDADDITIONALRESULTSFORTHECASEOFCO
12、NTINUOUSDEMANDVALUATIONS,ANDFOUNDTHATTHEREISABASISFORTHESELLERTOIMPROVEREVENUESBASEDONTHEEFFECTIVEDESIGNOFTHEGROUPBUYINGAUCTIONPRICECURVEDESIGNTHEORYTHEMODELFORTHEGROUPBUYINGAUCTIONMECHANISMWITHUNCERTAINBIDDERARRIVALTHATWEWILLDEVELOPSPANSTHREESTREAMSOFLITERATUREDEMANDUNCERTAINTY,CONSUMERBEHAVIORANDR
13、ELATEDMECHANISMDESIGNISSUESAUCTIONECONOMICSANDMECHANISMDESIGNTHEORYANDCURRENTTHEORETICALKNOWLEDGEABOUTTHEOPERATIONOFGROUPBUYINGAUCTIONSFROMTHEISANDELECTRONICCOMMERCELITERATUREDEMANDUNCERTAINTY,CONSUMERBEHAVIORANDMECHANISMDESIGNDEMANDUNCERTAINTIESTYPICALLYARECOMPOSEDOFCONSUMERDEMANDENVIRONMENTUNCERTA
14、INTYORUNCERTAINTYABOUTTHEAGGREGATELEVELOFCONSUMERDEMANDANDRANDOMNESSOFDEMANDINTHEMARKETPLACEREFLECTEDINBRIEFTEMPORALCHANGESANDDEMANDSHOCKSTHATARENOTEXPECTEDTOPERSISTCONSUMERUNCERTAINTYABOUTDEMANDINTHEMARKETPLACECANOCCURBASEDONTHEVALUATIONOFPRODUCTS,ANDWHETHERCONSUMERSAREWILLINGTOPAYHIGHERORLOWERPRIC
15、ESITMAYALSOOCCURONTHEBASISOFDEMANDLEVELS,ESPECIALLYTHENUMBEROFTHECONSUMERSINTHEMARKETFINALLY,THEREARETEMPORALCONSIDERATIONS,WHICHINVOLVEWHETHERACONSUMERWISHESTOBUYNOW,ORWHETHERTHEYMAYBESAMPLINGQUALITYANDPRICINGWITHTHEINTENTIONOFBUYINGLATERWEDISTINGUISHBETWEENDIFFERENTDEMANDLEVELENVIRONMENTSINADDITIO
16、N,ITISPOSSIBLETHATTHESECONSUMERDEMANDENVIRONMENTSMAYCOEXIST,ASISOFTENTHECASEWHENFIRMSMAKESTRATEGIESFORPRICEDISCRIMINATIONTHISPROMPTSASELLERTOCONSIDERSETTINGMORETHANONEPRICELEVEL,ASWEOFTENSEEINREALWORLDRETAILING,ASWELLASGROUPBUYINGAUCTIONSDANA2001POINTEDOUTTHATWHENAMONOPOLYSELLERFACESUNCERTAINTYABOUT
17、THECONSUMERDEMANDENVIRONMENT,ITUSUALLYWILLNOTBEINHISBESTINTERESTTOSETUNIFORMPRICESFORALLCONSUMERSTHEAUTHORSTUDIEDASCENARIOINWHICHTHEREWEREMOREBUYERSASSOCIATEDWITHHIGHDEMANDANDFEWERBUYERSASSOCIATEDWITHLOWDEMANDINTHEAUTHORSPROPOSEDPRICEMECHANISM,THESELLERSETSAPRICECURVEINSTEADOFASINGLEPRICE,SOASTOBEAB
18、LETOOFFERDIFFERENTPRICESDEPENDINGONTHEDIFFERENTDEMANDCONDITIONSTHATAPPEARTOOBTAININTHEMARKETPLACEITMAYBEUSEFULINSUCHSETTINGSTOEMPLOYANAUTOMATEDPRICESEARCHINGMECHANISM,WHICHISDEMONSTRATEDTOBEMOREROBUSTTOTHEUNCERTAINDEMANDTHANAUNIFORMPRICEMECHANISMWILL,RELATIVETOEXPECTEDPROFITSUNLIKEDANAS2001WORKTHOUG
19、H,WEWILLSTUDYSETTINGSINWHICHTHEREAREFEWERBUYERSWHOEXHIBITDEMANDATHIGHERPRICESANDMOREBUYERSWHOEXHIBITDEMANDATLOWERPRICESTHISISAUSEFULWAYTOCHARACTERIZEGROUPBUYING,SINCEMOSTPARTICIPATINGCONSUMERSTRULYAREPRICESENSITIVE,ANDTHISISWHATMAKESGROUPBUYINGAUCTIONINTERESTINGTOTHEMNOCKEANDPEITZ2007HAVESTUDIEDRATI
20、ONINGASATOOLTHATAMONOPOLISTTOOPTIMIZEITSSALESPOLICYINTHEPRESENCEOFUNCERTAINDEMANDTHEAUTHORSEXAMINEDTHREEDIFFERENTSELLINGPOLICIESTHATTHEYARGUEAREPOTENTIALLYOPTIMALINTHEIRENVIRONMENTUNIFORMPRICING,CLEARANCESALES,ANDINTRODUCTORYOFFERSAUNIFORMPRICINGPOLICYINVOLVESNOSELLERPRICEDISCRIMINATION,THOUGHCONSUM
21、ERSARELIKELYTOEXHIBITDIFFERENTLEVELSOFWILLINGNESSTOPAYWHENTHEYAREPERMITTEDTOEXPRESSTHEMSELVESTHROUGHPURCHASESATDIFFERENTPRICELEVELSACURRENTEXAMPLEOFUNIFORMPRICINGPOLICYISITUNESWWWITUNESCOM,WHICHHASBEENOFFERING99PERSONGPRICINGTHECONSUMERHASTODEALWITHVERYLITTLEUNCERTAINTYINTHEPROCESS,ANDTHISMAYBEAGOOD
22、APPROACHWHENTHESELLERWANTSTO“TRAIN”CONSUMERSTODEVELOPSPECIFICBUYINGHABITSASSEEMSTOHAVEBEENTHECASEWITHTHEONLINEPURCHASEOFDIGITALMUSICINTHEPASTFEWYEARSNOCKEANDPEITZ2007CHARACTERIZEDACLEARANCESALESPOLICYASCHARGINGAHIGHPRICEINITIALLY,BUTTHENLOWERINGTHEPRICEANDOFFERINGTHEREMAININGGOODSTOLOWVALUECONSUMERS
23、,ASISOFTENSEENINDEPARTMENTSTORESALESPOLICYCONSUMERSWITHAHIGHVALUATIONFORTHESALEGOODSMAYDECIDETOBUYATTHEHIGHPRICE,SINCETHEENDOGENOUSPROBABILITYOFRATIONINGBYTHESELLERISHIGHERATTHELOWERPRICEAPROPOSTOTHIS,CONSUMERSWHOBUYLATEATLOWPRICESTYPICALLYFINDTHATITISDIFFICULTTOFINDTHESTYLES,COLORSANDSIZESTHATTHEYW
24、ANT,ANDTHEYMAYHAVEMOREDIFFICULTYTOCOORDINATETHEPURCHASEOFMATCHINGITEMSEG,MATCHINGCOLORSANDSTYLESOFCLOTHINGINTRODUCTORYOFFERSCONSISTOFSELLINGALIMITEDQUANTITYOFITEMSATALOWPRICEINITIALLYINTHEMARKET,ANDTHENRAISINGPRICEAVARIANTOCCURSWHENTHESELLEROFFERSALOWERPRICEFORTHEFIRSTPURCHASEOFGOODSORSERVICESTHATTY
25、PICALLYINVOLVEMULTIPLEPURCHASESBYTHECONSUMEREG,BOOKCLUBMEMBERSHIPSANDCELLPHONESERVICESCONSUMERSWHOPLACEAHIGHVALUATIONONASALEITEMRATIONEDINITIALLYATTHELOWERPRICEMAYFINDITOPTIMALTOBUYTHEGOODSATTHEHIGHERPRICEINTRODUCTORYOFFERSMAYDOMINATEUNIFORMPRICING,BUTARENEVEROPTIMALIFTHESELLERUSESCLEARANCESALESINUN
26、CERTAINMARKETS,BUYERSWILLHAVEPRIVATEINFORMATIONCHEANDGALE2000POINTEDOUTTHATWHENCONSUMERSHAVEPRIVATEINFORMATIONABOUTTHEIRBUDGETCONSTRAINTSANDTHEIRVALUATIONOFSALESITEMS,SOAMONOPOLISTSOPTIMALPRICINGSTRATEGYISTOOFFERAMENUOFLOTTERIESONTHELIKELIHOODOFCONSUMERPURCHASESOFITSPRODUCTSATDIFFERENTPRICESANOTHERA
27、PPROACHISINTERTEMPORALPRICEDISCRIMINATIONBYOFFERINGDIFFERENTPRICESWITHDIFFERENTPROBABILITIESFORTHECONSUMERTOOBTAINTHEGOOD,THEMONOPOLISTCANPROFITABLYSEGMENTCONSUMERSEVENTHOUGHVALUATIONSEGMENTSALONEARENOTPROFITABLEEVENWHENTHESELLERCANEFFECTIVELYIDENTIFYTHECONSUMERDEMANDLEVELINTHEMARKETPLACE,DUETOSTOCH
28、ASTICFACTORSINTHEMARKETENVIRONMENT,ITSTILLMAYBEDIFFICULTFORTHESELLERTOEFFECTIVELYPREDICTDEMANDASARESULT,THESELLERMAYTRYTOIMPROVEITSDEMANDFORECASTBYUTILIZINGMARKETSIGNALSTHATMAYBEOBSERVEDWHENSALESOCCURHOWEVER,THEREARELIKELYTOBESOMESTOCHASTICDIFFERENCESBETWEENTHEPREDICTEDDEMANDBYTHESELLERANDTHEREALIZE
29、DDEMANDINTHEMARKETPLACEKAUFFMANANDMOHTADI2004LOANDWU2003POINTEDOUTTHATATYPICALSELLERFACESDIFFERENTTYPESOFRISKS,ANDAMONGTHESE,AKEYFACTORISFORECASTERROR,THEDIFFERENCEBETWEENTHEFORECASTANDTHEACTUALLEVELSOFDEMANDDIRIMANDROUNDY2002QUANTIFIEDFORECASTERRORSBASEDONASCHEMETHATESTIMATESTHEVARIANCEANDCORRELATI
30、ONOFFORECASTERRORSANDMODELSTHEEVOLUTIONOFFORECASTSOVERTIME22SOMEPROPERTIESOFGROUPBUYINGAUCTIONMECHANISMSOMEOFTHEKEYCHARACTERISTICSASSOCIATEDWITHGROUPBUYINGAUCTIONMECHANISMDESIGNAREPRESENTINTHELITERATURETHEGROUPBUYINGAUCTIONMECHANISMISFUNDAMENTALLYDIFFERENTFROMTHETYPICALQUANTITYDISCOUNTMECHANISMDOLAN
31、1987,WENG1995,CORBETTANDDEGROOTE2000THATISOFTENUSEDINCONSUMERANDBUSINESSTOBUSINESSPROCUREMENTSETTINGSFIRST,GROUPBUYINGCLOSINGPRICESTYPICALLYDECLINEMONOTONICALLYINTHETOTALPURCHASEQUANTITIESOFPARTICIPATINGBUYERS,ANDNOTJUSTBASEDONANINDIVIDUALBUYERSPURCHASEQUANTITIESSOAGROUPBUYINGAUCTIONDOESNOTLEADTOPRI
32、CEDISCRIMINATIONAMONGDIFFERENTBUYERSANDEVERYBUYERWILLBECHARGEDTHESAMECLOSINGPRICESECOND,INGROUPBUYINGAUCTIONS,IMPERFECTINFORMATIONMAYHAVEANIMPACTONPERFORMANCEANDMAKETHEFINALAUCTIONPRICEUNCERTAINGROUPBUYINGISNOTTHESAMEASWHATHAPPENSWITHCORPORATESHOPPINGCLUBSORAFFINITYGROUPBASEDBUYINGTHOUGHWITHTHESEOTH
33、ERMECHANISMS,CONSUMERSWILLBEASSOCIATEDWITHONEANOTHERINSOMEWAY,ANDBEABLETOOBTAINQUANTITYDISCOUNTSASARESULTANOTHERVARIANTOFTHEQUANTITYDISCOUNTMECHANISMOCCURSONTHEINTERNETWITHSHOPPINGCLUBSAND“POWERBUYING”WEBSITESNETMARKETCOMWWWNETMARKETCOM,BUYERSADVANTAGEWWWBUYERSADVANTAGECOM,ANDONLINECHOICEWWWONLINECH
34、OICECOMAREEXAMPLESTHATWEHAVERECENTLYOBSERVEDINTHEMARKETPLACEWITHUNCERTAINTYABOUTTHEULTIMATENUMBEROFTHEBIDDERSWHOWILLPARTICIPATE,INTERESTEDCONSUMERSMAYNOTKNOWWHETHERTHEYCANGETTHEPRODUCTS,ORWHATTHECLOSINGPRICEWILLBEWHENTHEYMAKEABIDTHISMAYEVENOCCURWHENTHEYBIDTHELOWESTPRICEONTHEGROUPBUYINGPRICECURVETHIR
35、D,INTHEQUANTITYDISCOUNTMECHANISM,TOACHIEVEADISCOUNTTHEBUYERMUSTORDERMORETHANTHETHRESHOLDNUMBEROFITEMSREQUIREDINGROUPBUYING,THEBUYERCANGETTHEDISCOUNTBYORDERINGMOREHERSELFORPERSUADINGOTHERBIDDERSTOORDERMORE,ASWESAWWITHTHE“TELLAFRIEND”LINKATLETSBUYITCOMFORCOBUYINGANDATTHEACTIVEGROUPDONATIONSITE,FUNDABL
36、ECOM,WWWFUNDABLECOMAFINALCONSIDERATIONINSOMEGROUPBUYINGAUCTIONSISTHATABUYERMAYBEABLETOCHOOSEHEROWNBIDDINGPRICE,WHICHMAKESTHISKINDOFAUCTIONSIMILARTOANOPENOUTCRYAUCTIONINPRACTICE,MANYBUYERSWILLONLYBEWILLINGTOSTATEALOWBIDPRICE,UNLESSTHEYCANRELYONTHEDESIGNOFTHEMECHANISMTOFAITHFULLYHANDLEINFORMATIONABOUT
37、THEIRACTUALRESERVATIONPRICEGROUPBUYINGAUCTIONSHAVEAKEY,BUTPARADOXICALFEATURETOREACHALOWERPRICEANDHIGHERSALEQUANTITYBUCKET,THECONSUMERMAYNEEDTOENTERTHEAUCTIONATAHIGHERPRICEANDLOWERSALESQUANTITYBUCKETCHENETAL2009出处JCHEN,RJKAUFFMAN,YLIU,XSONGSEGMENTINGUNCERTAINDEMANDINGROUPBUYINGAUCTIONSRELECTRONICCOMM
38、ERCERESEARCHANDAPPLICATIONS2009,3001二、翻译文章标题网络团购中的不确定性需求细分译文不确定性需求,是卖家通过网络拍卖形式销售产品的决策过程中的一个关键因素。我们从低价需求和高价需求的角度出发,在团购拍卖中探索不确定性需求。我们关注于一个垄断性团购零售商的分析,即向那些需要不同产品价值的消费者销售产品。同时,我们也检测团购买方在市场上会有何种表现,当团购买方面对具有竞争力标价的卖方时,卖方在类似于不确定性需求的假设下销售同种产品。以一个纳什投标者对这些卖方竞争结构的均衡分析策略为依据,我们能够描述出一种团购拍卖出价者占优策略。我们得到了许多有趣的发现。低价需求
39、中的团购很可能比在高价需求中的更有效。需求的结构问题。这项发现和新型相机、下一代卫星处理器、计算机和其他那些看似不太可能在团购市场上有效销售的高价产品有关。我们为了案例的连续需求能够有额外结果,发现有一个以团购拍卖有效价格为基础的可以改善卖方收入的设计曲线。关键词消费者行为,投标策略,需求不确定性,经济分析,电子市场,团购拍卖,市场机制,价前机制,模拟,不确定风险。IT产业先进的发展,令用创新商业模式处理业务问题成为可能。随着网络的日益发展,许多新的电子拍卖机制出现了,当一项拍卖的成本流失最小或无成本的时候,拍卖通常被理解为是为了设定比标价更高的预期卖方收入王1993。我们见过的一些新机制包括
40、美国在线、荷兰式拍卖、自命名和“现在买”机制,有一个易趣的荷兰式拍卖的例子为销售多个同类产品的描述。我们观察了另一种电子市场机制团购拍卖,一个同质多单位拍卖米切尔2002,李等。2004。基于网络的卖家和数字媒介已经接受了这种市场机制例如“LETSBUYIT”WWWLETSBUYITCOM和“EWINWINCOM”WWWEWINWINCOM,这些网站提供不同于传统拍卖的“转换市场”机制。在传统的拍卖中,竞拍的人为了成为胜利者,彼此间进行抗争。然而,在团购拍卖中,投标者用一个诱因来集合他们的出标,以至于卖方或数码调解人提供一个更低的价格,使他们全都能买到自己想要的产品赫恩以及其他人。2000。麦
41、长基以及其他人1991已经在实验研究探索了多数单位的成功拍卖。本篇论文是第一个探索不确定性需求在网络团购拍卖模式中的影响的论文。以一个纳什投标者为一个专营商卖方和一个竞争卖方的均衡分析战略为依据,我们能描述一个团购拍卖出价者对称占优的策略。我们发现网站上低价需求的团购很可能比高价需求的更具有效果。从而,产生了不同的消费者愿付水平需求结构问题。这和新型相机,下一代微处理器,计算机和其他高价值的产品市场有关系。我们为了案例的连续需求能够有额外结果,发现有一个以团购拍卖有效价格为基础的可以改善卖方收入的设计曲线。理论这个模型是有关于团购拍卖机制中的需求不确定者的,我们将涉及到三方面的文献跨度。不确定
42、性需求,消费者行为和涉及到机制设计问题;拍卖经济学和机制设计理论;当前IS和电子商务文学中的团购拍卖运行理论知识。需求不确定性,消费者行为和机制设计需求不确定性通常由消费需求环境的不确定性(或对消费需求总体水平的不确定性)和市场随机性需求(简言之,就是反映典型的持续变动和需求震动)组成的。消费的不确定性是关于以产品计价为依据的的市场需求,主要看消费者是否愿意付出更高或更低的价格。它也可能发生在以需求水平为依据尤其是市场消费者数目上。最终,有消费者是否愿意现在购买的暂时性考虑,或他们在购买后的质量和价格是否可抽样的长期考虑。我们在不同的需求水平环境中进行区分。此外,这些消费需求环境很可能共存,正
43、如工厂制定价格歧视策略也是常有的事情。这就使卖家考虑设定不止一个价格水平,正如我们经常在现实中看到零售,和团购拍卖一样。达纳2001指出当一个垄断卖方不确定消费需求环境时,通常不会对所有消费者设定一个统一的价格来求自身利益最大化。作者研究了一种情形,这种情形中有和高需求相关的较多买家和低需求相关的较少买家。在作者提出的价格机制中,卖方设定一个价格曲线而非单一价格,这样能根据在市场中出现的不同需求情况来提供不同的价格。它可以在这种环境中利用一个自动觅价机制,相对于期望的利润而言,需求不确定性已经被论证了比一个统一价格机制需求更坚固。然而不像达纳2001研究的那样,我们将会在较高价的少量买家的展会
44、需求、和在较低价的多数买家的展会需求的情况下进行研究。这有一个有用的方式来描述团购,多数参与的消费者对价格非常敏感,这是团购拍卖让他们感兴趣的原因所在。诺奇和派茨2007研究了作为专营商不确定性需求的一种工具,限量供应用以优化其销售政策。作者在销售环境中考察了三个不同销售政策,并试图找出哪一种是最佳的选择统一定价,清仓减价销售,厂家供应。一个统一的定价策略不包括卖方价格歧视,虽然消费者在被允许通过不同价格水平购买表达自己的时候,很可能显现出不同的愿意支付水平。当前一个统一定价策略的例子就是苹果公司开发的与IPOD搭配的在电脑上使用的音乐播放器和音乐管理器ITUNESWWWITUNESCOM,它
45、就是一直提供每歌曲99的价格。消费者不得不在过程中处理非常小的不确定性,当卖方想培养消费者开发具体的购买习惯时,这也许是一个好的途径,(就是在过去的几年中经常见到的从网上下载音乐)。诺奇和派茨2007阐述了一个清晰的销售政策,最初是制定高价,但是之后又降低价格,给低价消费者提供剩余产品经常在百货公司销售政策中见到的那样。高价值的消费者可能决定在销售产品高价时就购买,虽然卖家限量供应内在可能性的价格比低价稍稍高一些。就这点而言,稍后在低价上购买的消费者,通常发现找到自己想要的风格、颜色和样式比较困难,他们可能在协调匹配的收购商品上有更多的困难(例如衣服的颜色搭配和样式风格)。导言的提供是由最初市
46、场上低价的限量产品组成的,然后提高价格。当卖方在初次购买产品和服务提供更低价格时,变化就发生了,这通常涉及到消费者的多重购买(例如预定俱乐部会员资格和手机服务)。消费者在一个最初低价定量配给的规模产品上设定一个高价,发现这在高价购买产品上可能是理想的。引言的提供可能主导统一价格,但是如果卖家清晰地销售,效果就绝对不会理想了。在不确定的市场上,买方会有私人信息。澈和盖尔2000指出消费者有关于他们的预算限制和他们的计价项目价值的私人信息。所以一个专营商的最佳价格策略是提供一个抽奖的名单,类似于消费者购买它的不同价位的产品。另一个方式涉及到差别取价。通过为消费者提供不同概论的价格使其获得益处,专营
47、商有利地区分消费者,即使价值区分不仅仅是有利可图的。尽管当卖家能够在市场上有效地鉴定消费者需求水平,但是由于市场环境的随机因素,有效预测需求对卖方而言可能仍旧是困难的。因此,卖方可能试图通过利用市场信号来改善对消费者需求的预测,这样在销售的时候可能观察被观察到。然而,卖方有可能在一些随机差异之间预测消费者需求,并在市场上实现这种需求(考夫曼,莫泰迪,2004)。罗,吴2003指出一个典型的卖方面对不同类型的风险,在这些风险中,一个关键因素在于预测误差,预测间的差异和需求的实际水平。德拉姆和朗德2002以一个策划为依据来量化预测错误,估算预测误差的关联性并随着时间的推移建立起预测演化模型。团购拍
48、卖机制的一些可能性文献中呈现出了和团购拍卖机制设计相关联的一些关键性特征。团购拍卖机制和典型数量折扣机制根本不同多兰1987,翁1995,科尔韦特,德格罗特2000,前者经常在消费者和企业对企业电子商务采购环境中被使用。首先,团购结算价格通常在总共参与者的需求总量上单调地下降,而非单纯地只依据买方的购买量。这么一个团购拍卖不会在不同的消费者之间引发价格歧视,每一个消费者将会支付统一的最终价。其次,在团购拍卖中,不完全信息可能会在产品性能上有影响,并制作最终拍卖不确定价格。但是团购和公司购物俱乐部或类似的团体购买不同。同其他的一些机制,消费者将会在某种程度上相互关联,并最终能够获得数量折扣。另一
49、种数量折扣机制的变化发生在网络上,例如购物俱乐部、“购买力”网站、网络营销网站WWWNETMARKETCOM,买方优势WWWBUYERSADVANTAGECOM和在线选择WWWONLINECHOICECOM,这些都是我们最近在市场上观察的一些网站。同最终参与投标的人数不确定性有关,感兴趣的消费者可能不知道他们是否能够得到产品,或者当他们出标时会是什么样的临界价格。这可能就可能发生在他们在团购价格曲线的最低价格上出价的时候。再者,在数量折扣机制上,为了获得折扣,买方的数量必须超过要求团购的最低限。在团购中,买家能够通过自己订购更多产品或劝其他人订购刚多产品来获得折扣,正如我们在“大家买”网站上看到的一个用“邀请好友”的链接来壮大买方团体一样(在团购网站WWWFUNDABLECOM上也有类似情况)。在某些团购拍卖中,最后要考虑的是一个买方可能选择他自己报的价格有些类似公开喊叫拍卖了。实际上,许多购买者不仅愿意规定一个低的价格,而且他们能依赖机制设计,详实地掌握他们实际预定的价格信息。团购拍卖有一个关键但自相矛盾的特征即为了得到一个更低的价格和一个更高的销售数额,消费者可能需要在高价低量的时候进入拍卖陈和其他人,2009。出处电子商务考证和应用