青島市房地產(chǎn)市場預(yù)警系統(tǒng)建模及其實(shí)證研究
本文關(guān)鍵詞:青島市房地產(chǎn)市場預(yù)警系統(tǒng)建模及其實(shí)證研究,由筆耕文化傳播整理發(fā)布。
中國海洋大學(xué)
碩士學(xué)位論文
青島市房地產(chǎn)市場預(yù)警系統(tǒng)建模及其實(shí)證研究
姓名:周琦
申請學(xué)位級別:碩士
專業(yè):管理科學(xué)與工程
指導(dǎo)教師:張勤生
20080530
青島市房地產(chǎn)市場預(yù)警系統(tǒng)建模及其實(shí)tlF研究
青島市房地產(chǎn)市場預(yù)警系統(tǒng)建模及其實(shí)證研究
摘要
房地產(chǎn)業(yè)在整個國民經(jīng)濟(jì)體系中屬于先導(dǎo)性、基礎(chǔ)性產(chǎn)業(yè),處于主導(dǎo)產(chǎn)業(yè)地位,其存在著明顯的周期波動規(guī)律。起伏過大的波動與房地產(chǎn)經(jīng)濟(jì)的持續(xù)健康穩(wěn)定發(fā)展相矛盾,但是目前,我國房地產(chǎn)市場運(yùn)行機(jī)制不甚完善,還沒有形成合理、有序、競爭、高效的市場運(yùn)行體系,房地產(chǎn)市場存在信息傳遞不暢、信息數(shù)據(jù)失真、市場行情展示手段落后和市場交易網(wǎng)絡(luò)封閉等?系列問題。因此,研究房地產(chǎn)預(yù)警系統(tǒng),設(shè)置房地產(chǎn)預(yù)警指標(biāo)體系,系統(tǒng)、科學(xué)、準(zhǔn)確地確定房地產(chǎn)安全區(qū)域,成為有關(guān)決策部門亟需解決的重大現(xiàn)實(shí)問題,又是學(xué)術(shù)界需要深入研究的重大理論問題。
針對這種情況,本文提出了房地產(chǎn)市場預(yù)警系統(tǒng)模型研究,為促進(jìn)房地產(chǎn)業(yè)的健康發(fā)展提供一定的理論依據(jù)。并在此基礎(chǔ)上,對青島市房地產(chǎn)市場進(jìn)行了實(shí)證研究。
本文通過研究,得到的研究成果及研究結(jié)論主要有以下幾個方面:
(1)基于對國內(nèi)外房地產(chǎn)預(yù)警研究現(xiàn)狀的分析,歸納總結(jié)了房地產(chǎn)預(yù)警的基本概念、基本原則和基本方法。在對國內(nèi)原有各類房地產(chǎn)指標(biāo)體系進(jìn)行研究的基礎(chǔ)上,確定房地產(chǎn)預(yù)警指標(biāo)。
(2)房地產(chǎn)作為社會經(jīng)濟(jì)系統(tǒng)的一個子系統(tǒng),具有非線性復(fù)雜系統(tǒng)的特性。本文立足于解決房地產(chǎn)系統(tǒng)的非線性問題,建立更為先進(jìn)科學(xué)的房地產(chǎn)預(yù)警系統(tǒng),避免房地產(chǎn)市場的非常態(tài)波動,促進(jìn)房地產(chǎn)市場的持續(xù)、健康、穩(wěn)定發(fā)展。在現(xiàn)有研究的基礎(chǔ)上,系統(tǒng)地分析了房地產(chǎn)預(yù)警的特點(diǎn)及功能特征,對房地產(chǎn)預(yù)警過程中的關(guān)鍵預(yù)警指標(biāo)進(jìn)行了辨識、預(yù)測、診斷、監(jiān)測和控制,構(gòu)建了具有理論性和實(shí)踐性的房地產(chǎn)預(yù)警系統(tǒng),為解決房地產(chǎn)預(yù)警問題提供了依據(jù)。
(3)本文介紹的房地產(chǎn)預(yù)警系統(tǒng),以神經(jīng)網(wǎng)絡(luò)理論和房地產(chǎn)預(yù)警理論為基礎(chǔ),構(gòu)建了預(yù)警模型。利用神經(jīng)網(wǎng)絡(luò)在預(yù)測和模式識別領(lǐng)域的成熟運(yùn)用,重點(diǎn)探
青島市房地產(chǎn)市場頂警系統(tǒng)建模及其實(shí)證研究
討基于神經(jīng)網(wǎng)絡(luò)理論的房地產(chǎn)預(yù)警的模型和方法,并利用該模型形成了房地產(chǎn)市場預(yù)警體系。
(4)在建立了房地產(chǎn)預(yù)警指標(biāo)體系的基礎(chǔ)上,本文提出了基于LVQ—RBF神經(jīng)網(wǎng)絡(luò)的房地產(chǎn)預(yù)警模型,該模型克服了傳統(tǒng)預(yù)警方法的不足,具有高度的并行性和全局性,提高了房地產(chǎn)預(yù)警系統(tǒng)的非線性、自學(xué)習(xí)性、自適應(yīng)性及大規(guī)模并行分布知識處理的能力,具有較高的精確度和適用性。
(5)本文依據(jù)前期研究理論成果,對青島市房地產(chǎn)風(fēng)險預(yù)警進(jìn)行實(shí)證分析,依據(jù)技術(shù)可能、經(jīng)濟(jì)合理、操作可行等原則,最終形成綜合預(yù)警分析結(jié)論。預(yù)警分析的結(jié)果與青島房地產(chǎn)發(fā)展的實(shí)際情況基本吻合,表明本項(xiàng)研究所建立的房地產(chǎn)預(yù)警模型系統(tǒng)有效可行,理論分析充分,實(shí)用價值高,為指導(dǎo)和調(diào)控房地產(chǎn)市場提供了科學(xué)依據(jù)。
關(guān)鍵詞:房地產(chǎn);神經(jīng)網(wǎng)絡(luò):系統(tǒng)建模;學(xué)習(xí)矢量化;預(yù)警;預(yù)測【I
metsyboadgn’Q
青島市房地產(chǎn)市場預(yù)警系統(tǒng)建模及je實(shí)證研究
ModeIingforRealEstateForecastingandEarlYWarning
Systemandiitssl:mpEmlilrlicaIIReseaesearrchhIinUingdao
Abstraot
Realestateindustry,whichisplayingaleadingroleinthedevelopmentofnationaleconomy,showsmoreandmoreobviouslythecharacteristicofcyclefluctuation.Itsexcessivefluctuationcontradictswith也esustainable,healthy.andstabledevelopmentoftherealestateeconomy.Currently,thereisstillmuchworktodotoperfecttherealestatemarketoperationmechanismofChina,tOformrational,orderly,competitiveandefficientmarketoperationsystemandtosolvethecurrentproblemsoftheunsmoothtransferringofinformation,thedistortionofinformationdata,thebackwarddisplayingmeansofmarketquotationandtheblockingofthemarkettransactionnetwork.IthasbecomeahotpointintheacademiccyclesandarOI/Sesintensiveconcernsoftherelatedpolicy?makingapartmentstodoresearchonrealestateearlywarningsystem,tosetuptherealestateearlywarningindexsystemandsystematically,scientificallyandaccuratelydefinethesecureregionofrealestate.
Toaddressthisproblem,theresearchofearlywarningmodelsofrealestatemarketisputforwardinthisdissertationwhichwillmakeitstheoreticalcontributiontothepromotionofthehealthydevelopmentofrealestateindustry.Onthebaseofthisresearch,empiricalstudyisalsodoneontherealestatemarketofQingdao.
Basedontheresearchabove,themainfindingsandtheconclusionsarementionedasfollows:
Firstofall,basedontheanalysisofthecurrentresearchesathomeandabroad,thisdissertationsummarizesthebasicconcepts,principlesandmethodsofforecastingandearlywarningofrealestateeconomy.Onthebaseoftheanalysisofthedomesticrealestateindexsystems,thisdissertationselectsandfixesonitsindexesofearlywarning.
Secondly,asasubsystemofthesocialandeconomicsystem,realestateshowscomplexnon—linearcharacteristics.Thisdissertationaimsatsolvingthenonlinearproblemoftherealestatesystemandestablishingamoreadvancedandscientificearlywarningsysteminordertopreventtherealestatemarketfromtheabnormalfluctuationandtomaintainthesustainable,healthyandstabledevelopmentofthereallII
estatemarket.On山efoundationofthepresentresearch,thedissertationsystematicallyanalysesthecharacteristicsandfunctionsoftherealestateearlywarning,makeanidentification,prediction,diagnosis,monitoringandcontrolofthekeyindexintheprocessofrealestateearlywarningandconstructatheoreticallyandpracticallyfeasiblerealestateearlywarningsystemwhichprovidesthebasisforthesolutionoftherealestateearlywarningproblems.
Thirdly.basedonneuralnetworkstheoriesandrealestateearningwarningtheories,thisdissertationintroducesitsforecastingandearlywarningsystemanddevelopsitsmodelofforecastingandearlywarning.ByutilizingtheneuralnetworkswhichiSmaturelyappliedinthefieldofforecastandmodelrecognition,thedissertationputsitsemphasesontheresearchofthemodelsandmethodsofrealestateforecastingandearlywarning.Andbasedonthesemodelsthisdissertationdevelopsitsownforecastingandearlywarningsystemofrealestatemarket.
Fourthly,basedontherealestateearlywarningindexsystem,thisdissertationdevelopstheLVQ..RBFneuralnetworksmodelofforecastingandearlywarning.Withhighparallelism,globalsuperiority,accuracyandapplicabilitythismodelhasovercomethedeficiencyoftraditionalearlywarningmethodsandhighlyimprovestherealestateearlywarningsystem’Snon-linearity,selfstudyingability,selfadaptabilityandtheabilitytoprocesslarge—scaleconcurrentlydistributedknowledge.
Finally,basedontheprevioustheoreticalresearchfindings,thisdissertationempiricallyanalysesoftherealestatemarketinQingdaoandformsitscomprehensiveearlywarningtheoriesandmethods.SincetheconclusionoftheanalysisofearlywarninginthisdissertationisinaccordancewiththepracticaldevelopmentoftherealestateinQingdao,therealestateearlywarningsystemestablishedinthisresearchisprovedtobefeasible,withfulltheoryanalysisandgoodpracticalvalue,andprovidescientificfoundationforguidingandcontrollingtherealestatemarkets.
KeyWords:RealEstate,,NeuralNetworks,SystemModeling,LearningVectorQuantization,EarlyWarning,Forecasting
本文關(guān)鍵詞:青島市房地產(chǎn)市場預(yù)警系統(tǒng)建模及其實(shí)證研究,由筆耕文化傳播整理發(fā)布。
本文編號:132522
本文鏈接:http://www.lk138.cn/jingjilunwen/fangdichanjingjilunwen/132522.html