城市商业地产是城市商贸水平的空间集中体现，是城市商业发展繁荣程度的反映， 也是城市化发展水平的表现。近年来电子商务以其诸多优势吸引大批消费者，替代部分 实体市场，导致城市商业地产实体商铺遭受巨大冲击。在线下、线上融合的新消费环境 下，研究城市商业地产面积规模对城市土地资源的科学规划和商业资本的合理利用都有 重要意义。另外新型冠状病毒造成的疫情使得国内城市商业活动短期内基本停滞，造成 大量商铺关门歇业。合理的城市商业地产面积规模是城市规划部门与企业高层决策者急 需破解的重要课题。 本文通过梳理城市商业地产预测方法和商业地产文献，提出本文研究的商业地产研 究范围，包括零售、餐饮、娱乐等用于商业的房地产，不包括办公楼。以城市商业地产 面积规模为研究对象，分别以西部重要省会城市西安市和成都市为例进行实证研究，重 点探究电子商务因素对西部地区城市商业地产面积规模的影响。根据2000年至2018年 商业地产数据，将相关性分析与灰色关联度方法相结合，确定了城镇人口、恩格尔系数、 地区生产总值、第三产业产值、商业营业用房投资额、社会消费品零售总额、城乡居民 储蓄存款、城镇居民人均可支配收入以及电子商务交易额和互联网用户规模10个与城 市商业地产相关性较高的指标体系，进而利用多指标BP神经网络，构建了新增城市商 业地产面积规模预测模型。 基于西安与成都商业地产发展现状，2000年至2018年，西安市新增商业地产面积 整体呈上升趋势，增幅较快，年增长率为26%；成都市呈波动增长趋势，增速相对较慢， 年增长率为14%。成都市年新增出售商业地产面积均高于西安市，每年新增出售规模差 距在28—140万平方米之间。根据BP神经网络模型预测，发现电子商务因素对城市商 业地产面积规模影响较大。2019年至2025年，引入电子商务因素前后，西安年新增出 售商业地产面积规模都呈平缓增长趋势，且每年新增出售面积差距保持在10—31万平 方米之内；未引入电商因素时商业地产面积规模和增速都大于引入后的面积规模，其中， 未引入电商因素的年增长率为3.65%，引入后年增长率为2.75%；成都年新增出售商业 地产面积规模呈倒“U”发展趋势，引入电商因素前后每年新增面积规模差距在15—60 万平方米之间；未引入电商因素预测的面积大于引入后的面积规模，但增速明显小于引 入后，其中，未引入电商因素的年增长率为4.94%，引入后年增长率为7.06%。预测结 果表明，2019年至2025年，成都市每年新增出售商业地产面积规模和整体增速都大于 ii 西安市。未引入电商因素时成都每年新增出售面积比西安市高64—200万平方米，引入 后成都每年新增出售面积比西安高113—200万平方米。 通过对比我国西部两个省会城市西安市和成都市商业地产面积规模在未来短期的 需求量发现，未来西部城市商业地产面积规模还将持续增加，达到一定规模后在2025 年开始出现下降趋势，但规模总体少于未引入电子商务影响下的需求量，网购消费的介 入使得城市商业地产面积规模明显减少。 关键词：商业地产，面积规模，电子商务，BP神经网络，预测 iii Abstract Urbancommercialrealestateistheconcentratedembodimentofurbancommerciallevel ofspace,reflectstheprosperityofurbancommercialdevelopment,butalsotheperformance ofurbanizationdevelopmentlevel.Inrecentyears,e-commerceattractsalargenumberof consumerswithitsmanyadvantagesandreplacesthemarketofsomephysicalcommodities. Electroniccommerceinrecentyears,withitsmanyadvantagestoattractalargenumberof consumers,andreplacethepartofthephysicalcommoditymarkets,particularlyintheretail industryoccupiestheimportantposition,leadtohitcitycommercialrealestateentitiesshops, andwiththeemergenceof"newretailsalesmodel",onlinefusionintegrationofmodern logisticsdevelopmentwillbemoreadaptabletomarketneeds.Inthenewconsumption environment,itisofgreatsignificancetostudythedevelopmentofcommercialrealestate areascalerequiredbythecityinthefuturefortheplanningofurbanlandresourcesandthe utilizationofsocialcapital.Inaddition,theoutbreakcausedbynovelcoronaviruspneumonia causedthedomesticurbancommercialactivitiestobasicallystopinashortperiodoftime, andthereasonablesizeoftheurbancommercialrealestateareahasbecomeanimportant issueurgentlyneededtobesolvedbytheurbanplanningdepartmentandthehigh-level decisionmakersofenterprises. Bycombingandsummarizingthepredictionmethodsandcommercialrealestate literatureofurbancommercialrealestate,thispaperproposestheresearchscopeof commercialrealestateinvolvedinthispaper,thatis,thenarrowcommercialrealestate, includingretail,wholesale,catering,entertainmentandothercommercialservices,excluding officebuildings.Takingtheareascaleofurbancommercialrealestateastheresearchobject, Xi'anandChengdu,twoimportantprovincialcapitalsinwesternChina,weretakenas examplestoconductempiricalresearch,focusingontheinfluenceofe-commerceandother factorsontheareascaleofurbancommercialrealestateinwesternChina.Accordingtothe datafrom2000to2018,commercialrealestate,combiningcorrelationanalysisandgrey relationalanalysis,determinestheurbanpopulation,GDP,tertiaryindustryoutputvalue, engelcoefficient,commercialbuildingsforbusinessinvestment,totalretailsalesofsocial consumergoods,urbanandruralresidentssavings,percapitadisposableincomeofurban iv resident,ande-commercetransactionsandthenumberofInternetusers10highcorrelation withthecitycommercialrealestateindexsystem,thenusingmulti-indexBPneuralnetwork, buildthesizeoftheurbanareaofcommercialrealestateshort-termpredictionmodelinthe future.TheannualincreaseofcommercialrealestatesoldinChengdushowsaninverted"U" developmenttrend.Theannualincreaseofcommercialrealestatesizebeforeandafterthe introductionofe-commerceisbetween150,000and600,000squaremeters.Thepredicted areawithoute-commercefactorislargerthantheareascaleaftertheintroduction,butthe growthrateisobviouslysmallerthanthataftertheintroduction BasedonthedevelopmentstatusofcommercialrealestateinXi'anandChengdu,from 2000to2018,thenewlyaddedcommercialrealestateareainXi'anhasbeenontherisewith anannualgrowthrateof26%.Chengdushowsatrendoffluctuatinggrowth,whichis relativelyslow,withanannualgrowthrateof14%.Theareaofcommercialrealestatenewly soldinChengduishigherthanthatinXi'an,andthegapofnewlysoldscaleisbetween 280,000and1.4millionsquaremeters.Throughcalculationandsimulation,itisfoundthat e-commercefactorshaveagreatinfluenceontheareascaleofurbancommercialreal estate.From2019to2025,beforeandaftertheintroductionofe-commerce,theannual increaseintheareaofcommercialrealestatesoldinXi'anshowsaflatgrowthtrend,andthe gapoftheannualincreaseintheareasoldremainswithin100,000-310,000squaremeters. Whenthefactorofe-commerceisnotintroduced,theareascaleandgrowthrateof commercialrealestatearebothlargerthantheareascaleintroduced.Amongthem,theannual growthratewithouttheintroductionofe-commerceis3.65%,andtheannualgrowthrateafter theintroductionofe-commerceis2.75%.Amongthem,theannualgrowthratewithoutthe introductionofe-commerceis4.94%,andtheannualgrowthrateaftertheintroductionof e-commerceis7.06%.Theforecastresultsshowthatfrom2019to2025,theareasizeandthe overallgrowthrateofnewcommercialrealestatesalesinChengduwillbehigherthanthatin Xi'an.Whenthee-commercefactorisnotintroduced,theannualnewlysoldareainChengdu is640,000to2millionsquaremetershigherthanthatinXi'an,andtheannualnewlysold areainChengduis1.13millionto2millionsquaremetershigherthanthatinXi'an. Bycomparingthetwowesternprovincialcapitalcityofxi'anandChengduareathesize v ofthecommercialrealestateshort-termdemandfoundinthefuture,thefuturescaleof commercialrealestateareainwesterncitieswillcontinuetoincrease,afterreachingacertain scale,in2025startedtodecline,buttheoverallscaleislessthanthedemandunderthe influenceoftheintroductionofe-commerce,onlineconsumerinvolvementmakescity commercialrealestateareascaledecreasedsignificantly. Keywords:Commercialrealestate，Areascale，ElectronicCommerce，BPneural network，Forecast vi 目录 第一章绪论...1 1.1研究背景.............................1 1.2研究目的与意义.................2 1.2.1研究目的..............2 1.2.2研究意义..............2 1.3研究思路与内容.................3 1.3.1研究思路..............3 1.3.2研究内容..............6 1.4研究创新点.........................6 第二章相关理论与文献综述......8 2.1城市商业地产定义.............8 2.2预测方法.............................8 2.3文献综述...........................12 2.3.1国外研究............12 2.3.2国内研究............13 2.4本章小结...........................15 第三章城市商业地产面积规模预测模型构建.......16 3.1城市商业地产面积规模影响因素指标体系确定.........................16 3.1.1影响因素概述....16 3.1.2影响因素指