基于互联网大数据的实时房屋租赁价格指数编制研究——以济南、青岛为例
|
摘要
为实时反映房屋租赁价格的动态演化特征,从房屋租赁价格形成的影响因素及其传导机制入手,借鉴最佳指数编制理论的思想,运用网络爬虫技术自动化采集与解析开源互联网房屋租赁挂牌交易大数据,筛选出适用于开源数据模式的最佳组合指数编制方法:Hedonic-Fisher链式指数模型,并利用济南、青岛两市2017年12月至2018年3月房屋租赁交易经验数据,实证分析了该方法的有效性且进行了指数质量检验。研究表明:该实时房屋租赁价格指数能够揭示房产属性变量与房屋租赁价格之间的异质关系,相比于双加权平均法和中位数价格法具有较好的计算质量;与传统的月度房地产价格指数不同的是该指数具有较强的实时性,领先70个大中城市商品住宅销售价格指数约35～45天,能够先行准确地反映租赁价格总水平的变动;指数的实时性从可视化的角度检验了宏观经济政策的微观效度:经验数据检验前期,济青两地房屋租赁价格水平较为稳定,随着2018年初"租购并举"住房制度的提出,经验数据检验后期,济青两市受政策影响房屋租赁价格上涨明显,所以实时房屋租赁价格呈现出先平稳后适度上涨的走势。
In order to reflect the dynamic evolution characteristics of housing rental price in real time, this paper, by starting with the influencing factors of housing rental price formation and its transmission mechanism, drawing on the idea of the best index compilation theory and adopting the network crawler technology to automatically collect and analyze the big data of open source Internet housing rental listing transactions, selects the best compilation method for the open source data model, i.e., the Hedonic-Fisher chain index model, and empirically analyzes the validity of this method and conducts an index quality test by adopting the empirical data of housing leasing transactions in Jinan and Qingdao from December 2017 to March 2018. The results show that the real-time housing rental price index can reveal the heterogeneous relationship between housing property variables and rental prices and has better calculation quality than the double-weighted average method and the median price method; unlike the traditional monthly real estate price index, this index has a strong real-time performance and leads the sales price index of commercial residential buildings in 70 large and medium-sized cities by about 35-45 days, accurately reflecting the changes of the total rental price level in advance; and that the real-time nature of the index examines the micro-validity of the macroeconomic policy from the visual point of view: the rental price level in both Jinan and Qingdao is relatively stable in the early period of the empirical data test while in the later period of the empirical data test and with the "rent-purchase" housing system introduced in early 2018, the rental price in both Jinan and Qingdao has increased significantly affected by the policy. Therefore, real-time housing rental prices present a trend of a moderate rise after stabilization.
In order to reflect the dynamic evolution characteristics of housing rental price in real time, this paper, by starting with the influencing factors of housing rental price formation and its transmission mechanism, drawing on the idea of the best index compilation theory and adopting the network crawler technology to automatically collect and analyze the big data of open source Internet housing rental listing transactions, selects the best compilation method for the open source data model, i.e., the Hedonic-Fisher chain index model, and empirically analyzes the validity of this method and conducts an index quality test by adopting the empirical data of housing leasing transactions in Jinan and Qingdao from December 2017 to March 2018. The results show that the real-time housing rental price index can reveal the heterogeneous relationship between housing property variables and rental prices and has better calculation quality than the double-weighted average method and the median price method; unlike the traditional monthly real estate price index, this index has a strong real-time performance and leads the sales price index of commercial residential buildings in 70 large and medium-sized cities by about 35-45 days, accurately reflecting the changes of the total rental price level in advance; and that the real-time nature of the index examines the micro-validity of the macroeconomic policy from the visual point of view: the rental price level in both Jinan and Qingdao is relatively stable in the early period of the empirical data test while in the later period of the empirical data test and with the "rent-purchase" housing system introduced in early 2018, the rental price in both Jinan and Qingdao has increased significantly affected by the policy. Therefore, real-time housing rental prices present a trend of a moderate rise after stabilization.
引文
[1]廖俊平.租购并举给房地产市场带来根本性变化[J].中国房地产,2018(1):78-79.
[2]张源.基于拉姆齐模型和房价租售比的上海市房地产租赁市场实证分析[J].统计科学与实践,2017(1):14-18.
[3]DEFUSCO A A,DING W J,FERNANDO F V,et al.The role of price spillovers in the American housing boom[J].Journal of Urban Economics,2018,108:72-84.
[4]徐强.国际视野下指数研究与实践的进展、动向与挑战[J].统计研究,2017,34(2):110-128.
[5]CAVALLO A.Online and official price indexes:measuring argentina's inflation[J].Journal of Monetary Economics,2013,60(2):152-165.
[6]顾光同,许冰.浅谈互联网大数据下房地产价格指数编制[J].中国统计,2017(10):31-33.
[7]孙易冰,赵子东,刘洪波.一种基于网络爬虫技术的价格指数计算模型[J].统计研究,2014,31(10):74-80.
[8]陈柯,韩清,孟美侠,等.一种新的利用网络爬虫技术的土地价格指数编制方法[J].数量经济技术经济研究,2017,34(3):128-144.
[9]张宏斌,贾生华.编制城市房地产价格指数的理论模型和实用方法[J].中国软科学,2000(4):65-68.
[10]OSLAND L.The importance of unobserved attributes in hedonic house price models[J].International Journal of Housing Markets and Analysis,2013,6(1):63-78.
[11]杨晓冬,李忠富.住宅租赁特征价格研究---以哈尔滨市为例[J].系统管理学报,2012,21(4):486-493.
[12]李素文.中国一线城市房租形成机理和调控对策[J].财经理论与实践,2014,35(6):114-118.
[13]田金方.消费者信心影响中国家庭消费的实证诠释[J].经济与管理评论,2013,29(2):76-81.
[14]刘琪.浅析地价与地租、房价与房租之间的关系[J].价格理论与实践,2006(8):41-42.
[15]FREEMAN A.M,HEMIGES J A,KLING P C L.The measurement of environmental and resource values:theory and methods(2ed Edition)[M].Washington D.C.:Resources for the Future,2003.
[16]MARSHALL K L,MCSWEENY S M,VILLACORTA A.Method and system for providing a home data index model:US,US20130103459[P].2013.
[17]李国柱,孙焕民.编制城市房地产价格指数的若干问题[J].统计研究,2003(8):28-30.
[18]孙玉环.基于海量交易数据的房地产特征价格模型的构建[J].统计与决策,2011(2):9-13.
[19]CASE B,QUIGLEY J M.The dynamics of real estate prices[J].Review of Economics&Statistics,1991,73(1):50-58.
[21]徐强.国际视野下指数研究与实践的进展、动向与挑战[J].统计研究,2017(2):2-9
[22]徐强.价格指数编制中的Hedonic质量调整方法研究[J].财经问题研究,2009(8):22-28.
[24]温海珍.城市住宅的特征价格:理论分析与实证研究[D].杭州:浙江大学,2004.
[25]BOKHARI S,GELTNER D.Estimating real estate price movements for high frequency tradable indexes in a scarce data environment[J].Journal of Real Estate Finance&Economics,2012,45(2):522-543.
[26]PEDERSEN T M.The hodrick-prescott filter,the slutzky effect,and the distortionary effect of filters[J].Journal of Economic Dynamics and Control,2001,25(8):1081-1101.
[27]郑思齐,孔鹏,郭晓旸.类重复交易房价指数编制方法与应用[J].统计研究,2013,30(12):41-47.
(1)若一个指数是某种灵活的函数(如效用函数、生产函数、单位成本函数等) 的精确反映,这一指数就称为最佳指数,最佳价格指数可以用来精确地估计理论上的真实生活费用指数。
[2]张源.基于拉姆齐模型和房价租售比的上海市房地产租赁市场实证分析[J].统计科学与实践,2017(1):14-18.
[3]DEFUSCO A A,DING W J,FERNANDO F V,et al.The role of price spillovers in the American housing boom[J].Journal of Urban Economics,2018,108:72-84.
[4]徐强.国际视野下指数研究与实践的进展、动向与挑战[J].统计研究,2017,34(2):110-128.
[5]CAVALLO A.Online and official price indexes:measuring argentina's inflation[J].Journal of Monetary Economics,2013,60(2):152-165.
[6]顾光同,许冰.浅谈互联网大数据下房地产价格指数编制[J].中国统计,2017(10):31-33.
[7]孙易冰,赵子东,刘洪波.一种基于网络爬虫技术的价格指数计算模型[J].统计研究,2014,31(10):74-80.
[8]陈柯,韩清,孟美侠,等.一种新的利用网络爬虫技术的土地价格指数编制方法[J].数量经济技术经济研究,2017,34(3):128-144.
[9]张宏斌,贾生华.编制城市房地产价格指数的理论模型和实用方法[J].中国软科学,2000(4):65-68.
[10]OSLAND L.The importance of unobserved attributes in hedonic house price models[J].International Journal of Housing Markets and Analysis,2013,6(1):63-78.
[11]杨晓冬,李忠富.住宅租赁特征价格研究---以哈尔滨市为例[J].系统管理学报,2012,21(4):486-493.
[12]李素文.中国一线城市房租形成机理和调控对策[J].财经理论与实践,2014,35(6):114-118.
[13]田金方.消费者信心影响中国家庭消费的实证诠释[J].经济与管理评论,2013,29(2):76-81.
[14]刘琪.浅析地价与地租、房价与房租之间的关系[J].价格理论与实践,2006(8):41-42.
[15]FREEMAN A.M,HEMIGES J A,KLING P C L.The measurement of environmental and resource values:theory and methods(2ed Edition)[M].Washington D.C.:Resources for the Future,2003.
[16]MARSHALL K L,MCSWEENY S M,VILLACORTA A.Method and system for providing a home data index model:US,US20130103459[P].2013.
[17]李国柱,孙焕民.编制城市房地产价格指数的若干问题[J].统计研究,2003(8):28-30.
[18]孙玉环.基于海量交易数据的房地产特征价格模型的构建[J].统计与决策,2011(2):9-13.
[19]CASE B,QUIGLEY J M.The dynamics of real estate prices[J].Review of Economics&Statistics,1991,73(1):50-58.
[21]徐强.国际视野下指数研究与实践的进展、动向与挑战[J].统计研究,2017(2):2-9
[22]徐强.价格指数编制中的Hedonic质量调整方法研究[J].财经问题研究,2009(8):22-28.
[24]温海珍.城市住宅的特征价格:理论分析与实证研究[D].杭州:浙江大学,2004.
[25]BOKHARI S,GELTNER D.Estimating real estate price movements for high frequency tradable indexes in a scarce data environment[J].Journal of Real Estate Finance&Economics,2012,45(2):522-543.
[26]PEDERSEN T M.The hodrick-prescott filter,the slutzky effect,and the distortionary effect of filters[J].Journal of Economic Dynamics and Control,2001,25(8):1081-1101.
[27]郑思齐,孔鹏,郭晓旸.类重复交易房价指数编制方法与应用[J].统计研究,2013,30(12):41-47.
(1)若一个指数是某种灵活的函数(如效用函数、生产函数、单位成本函数等) 的精确反映,这一指数就称为最佳指数,最佳价格指数可以用来精确地估计理论上的真实生活费用指数。