| 1,160 | 14 | 106 |
| 下载次数 | 被引频次 | 阅读次数 |
传统空间计量模型采用单一不变系数表征单元间的空间关系,但随着样本中个体和时间的增大,不变系数模型难以准确反映空间关系的时变性,可能导致参数估计有偏。基于此,本文构建时变系数广义空间滞后模型,利用贝叶斯方法和MCMC抽样估计模型参数,并与不变系数模型作对比,最后应用于具体实例。数值模拟结果表明,时变系数模型参数估计的平均偏差和均方误根都小于不变系数模型,且均方误根随个体或时间的增大而减小。实例应用不仅重新测度了产业集聚对产业结构升级影响的空间时变效应,还再次证实了模型和方法的实用性。
Abstract：Traditional spatial econometric models use a single invariant coefficient to characterize the spatial relationship between the units. However,with the increase of individuals and time in the sample,the invariant coefficient model cannot accurately reflect the time variability of the spatial relationship,which may lead to biased parameter estimates. Based on this,this paper constructs a generalized spatial lag model with time-varying coefficients,uses Bayesian method and MCMC sampling to estimate model parameters,compares it with invariant coefficient model,and finally applies it to specific examples. The numerical simulation results show that the mean deviation and root mean square error of the parameter estimates of the time-varying coefficient model are smaller than those of the invariant coefficient model,and the root mean square error decreases with the increase of individuals or time. The example application not only re-measures the timevarying effect of the impact of industrial agglomeration on the upgrading of industrial structure,but also confirms the practicability of the model and method.
[1]陈建宝，孙林．随机效应变系数空间自回归面板模型的估计[J]．统计研究，2017,34(5):118-128.
[2]邓明．时变系数的空间误差合成模型———基于FGLS和GM的多阶段迭代估计[J]．数量经济技术经济研究，2013,30(4):111-123.
[3]邓明．时变系数空间自回归面板数据模型的极大似然估计[J]．统计研究，2016,33(9):96-103.
[4]邓明，钱争鸣．我国省际知识生产及其空间溢出的动态时变特征———基于Spatial SUR模型的经验分析[J]．数理统计与管理，2013,32(4):571-585.
[5]纪玉俊，李超．创新驱动与产业升级———基于我国省际面板数据的空间计量检验[J]．科学学研究，2015,33(11):1651-1659.
[6]李虹，邹庆．环境规制、资源禀赋与城市产业转型研究———基于资源型城市与非资源型城市的对比分析[J]．经济研究，2018,53(11):182-198.
[7]藤田昌久，克鲁格曼，维纳布尔斯．空间经济学:城市区域与国际贸易[M]．梁琦译，北京:中国人民大学出版社，2005.
[8]徐朝阳．工业化与后工业化:“倒U型”产业结构变迁[J]．世界经济，2010,33(12):67-88.
[9]徐敏，姜勇．中国产业结构升级能缩小城乡消费差距吗?[J]．数量经济技术经济研究，2015,32(3):3-21.
[10]Anselin L. Spatial Econometrics:Methods and Models[J].Journal of the American Statistical Association,1988,85(411):905-907.
[11]Aquaro M,Bailey N,Pesaran M H. Quasi Maximum Likelihood Estimation of Spatial Models with Heterogeneous Coefficients[J].USC-INET Research Paper,2015.
[12]Blasques F,Koopman S J,Lucas A,et al. Spillover Dynamics for Systemic Risk Measurement Using Spatial Financial Time Series Models[J]. Journal of Econometrics,2016,195(2):211-223.
[13]Brunsdon C,Fotheringham A S,Charlton M E. Geographically Weighted Regression:A Method for Exploring Spatial Nonstationarity[J]. Geographical Analysis,1996,28(4):281-298.
[14]Cai Z. Trending Time-varying Coefficient Time Series Models with Serially Correlated Errors[J]. Journal of Econometrics,2007,136(1):163-188.
[15]Catania L,Bille A G. Dynamic Spatial Autoregressive Models with Autoregressive and Heteroskedastic Disturbances[J]. Journal of Applied Econometrics,2017,32(6):1178-1196.
[16]Chenery H,Syrquin M,Elkington H. Patterns of Development:1950-1970[M]. London:Oxford University Press,1975.
[17]Crespo R,Fotheringham S,Charlton M. Application of Geographically Weighted Regression to A 19-year Set of House Price Data in London to Calibrate Local Hedonic Price Models[C]. Proceedings of the 9th International Conference on Geocomputation. National University of Ireland Maynooth,2007.
[18]Elhorst J P. Specification and Estimation of Spatial Panel Data Models[J]. International Regional Science Review,2003,26(3):244-268.
[19]Fan J,Huang T. Profile Likelihood Inferences on Semiparametric Varying-coefficient Partially Linear Models[J]. Bernoulli,2005,11(6):1031-1057.
[20]Kelejian H H,Prucha I R. A Generalized Moments Estimator for the Autoregressive Parameter in a Spatial Model[J]. International Economic Review,1999,40(2):509-533.
[21]Lee L F. Asymptotic Distributions of Quasi-maximum Likelihood Estimators for Spatial Autoregressive Models[J]. Econometrica,2004,72(6):1899-1925.
[22]Le Sage J P,Chih Y Y. A Bayesian Spatial Panel Model with Heterogeneous Coefficients[J]. Regional Science and Urban Economics,2018,72:58-73.
[23]Le Sage J P. Spatial Econometric Panel Data Model Specification:A Bayesian Approach[J]. Spatial Statistics,2014,9:122-145.
[24]Li D,Chen J,Gao J. Nonparametric Time-varying Coefficient Panel Data Models with Fixed Effects[J]. The Econometrics Journal,2011,14(3):387-408.
[25]Li K. Spatial Panel Data Models with Structural Change[J]. MPRA Paper,2018.
[26]Pace R K,Le Sage J P,Zhu S. Spatial Dependence in Regressors and Its Effect on Performance of Likelihood-based and Instrumental Variable Estimators[J]. Advances in Econometrics,2012,30:257-295.
[27]Sengupta A. Testing for a Structural Break in a Spatial Panel Model[J]. Econometrics,2017,5(1):1-17.
[28]Smith T E,Le Sage J P. A Bayesian Probit Model with Spatial Dependencies[J]. Advances in Econometrics,2004,18(18):127-160.
[29]Su L,Jin S. Profile Quasi-maximum Likelihood Estimation of Partially Linear Spatial Autoregressive Models[J]. Journal of Econometrics,2010,157(1):18-33.
[30]Su L. Semiparametric GMM Estimation of Spatial Autoregressive Models[J]. Journal of Econometrics,2012,167(2):543-560.
(1)因篇幅所限,时变系数模型各参数详细推导过程以附录1展现。
(1)另一种参数更新方式是先从参数条件后验分布中抽样更新T期(a_t~1β_t~1,(σ~2)_t~1),t=1,…,T,再基于所有新的参数取值(a~1,β~1,(σ~2)~)更新得到参数(φ_t~1,ρ_t~1),t=1,…,T,对上述过程重复进行直到达到最大循环次数。
(1)Rook相邻空间权重矩阵中若两区域有共同的边,则w_(ij)=1,否则w_(ij)=0。
(2)Case相邻空间权重矩阵中,假设共有R个区域,每个区域有M个个体,每个个体被赋予相同权重,则W=I_R*B_M总样本数W=R×M其中B_M=(l_Ml'_M-I_M)/(M-1),I_M是M×1维全为1的列向量。
(3)因篇幅所限,文中另一种稳健性说明是对Rook相邻下QML方法与贝叶斯方法的模拟用时和模拟结果进行比较,发现贝叶斯方法比QML方法的模拟用时更短、模拟效果更好,以附录2展示。
(4)因篇幅所限,不变系数模型各参数条件后验分布及抽样步骤以附录1展现。
(1)三期叠加是指增长速度换挡期、结构调整阵痛期、前期刺激政策消化期。
(1)由于海南省三沙市、海南省詹州市、贵州省毕节市、贵州省铜仁市、西藏自治区拉萨市、西藏自治区日喀则市、青海省海东市存在大量数据缺失,放上述各市未列入研究范围。
(1)参照空间杜宾模型的参数估计原理,解释变量滞后项的系数与解释变量系数的估计方法相同。
(2)不变系数模型的实证回归结果基于附录1中的抽样步骤。
(3)参考中国社科院发布的《2012年经济蓝皮书》。
基本信息:
DOI：10.19343/j.cnki.11-1302/c.2020.11.010
中图分类号:O212.8
引用信息:
[1]陶长琪,徐茉.时变系数广义空间滞后模型的贝叶斯估计[J].统计研究,2020,37(11):116-128.DOI:10.19343/j.cnki.11-1302/c.2020.11.010.
基金信息:
国家社会科学基金重大项目“高质量发展视阈下创新要素配置的统计测度与评价研究”（19ZDA121）;; 国家自然科学基金项目“制造业高质量发展视阈下创新要素的再配置机理及优化策略研究”（71973055）;国家自然科学基金项目“区际产业梯度转移与升级中的技术势能集聚、转换及空间效应研究”（71773041）
2020-10-23
2020-10-23
2020-10-23