文档介绍：华中科技大学
硕士学位论文
面板协整模型及其在中部城镇居民消费中的应用
姓名:汪萍
申请学位级别:硕士
专业:数量经济学
指导教师:王少平
20080528
华中科技大学硕士学位论文
摘要
面板数据模型是近年来计量经济学研究的热点之一,该模型能够同时在时间和
截面单元两个方面反映研究对象的特点,因而具有单纯时间序列模型和横截面模型
所不具备的优点。随着计量经济学理论及应用的发展,越来越多的时候需要同时采
用横截面和时间序列数据。因此,面板数据模型得到了广泛的应用。
在理论阐述部分,本文首先简单的介绍了面板数据模型,然后详细介绍了几种
最新的面板单位根检验和面板协整检验以及面板协整式的估计,其中的重点是面板
动态最小二乘法。
在实证研究部分,本文主要研究的是中部城镇居民的消费问题,采用了时间序
列单位根检验和面板单位根检验。在检验结果为存在单位根的基础上,又对数据进
行了时间序列协整检验和面板协整检验,检验结果显示:中部六省城镇居民生活性
消费和可支配收入间以及二者取自然对数所得序列间均存在面板协整。之后采用面
板 DOLS 方法来估计两个面板协整式,得到中部城镇居民的边际消费倾向和消费收
入弹性。
在最后的结论部分中,本文进一步讨论了检验和估计的结果,指出了中部城镇
居民消费中存在的问题,并结合实际提出了可行的解决办法和政策建议。
关键词:面板数据模型; 面板单位根检验; 面板协整检验; 面板动态最小二乘法
I
华中科技大学硕士学位论文
Abstract
Nowadays, panel data models is increasingly popular in the field of Econometrics,
which could reflect characteristics of a particular object in two aspects-- times series and
cross-sections, thus have the merits which the pure time series models and the
cross-sections models do not have. With the development of the theory and application in
Econometrics, both of the times series data and cross-sections data are needed, and
therefore the panel data models are widely used.
In the theory section, this paper briefly introduces panel data models, and then
introduces several latest panel unit root tests and panel cointegration tests, and the ways
to estimate the cointegration equation. Panel DOLS is the pivot of this section.
In the application section, this paper mainly studies the consumption of city
dwellers of the six provinces in central China, using time series unit root tests and panel
data unit root tests. As the result is that there are unit roots and panel unit roots in the
consumption series and e series and their hyperbolic logarithm series, the next step
is to do cointegration tests. And then we get that both of the two couples have the relation
ship of cointegration. Then the task is to estimate the cointegration equations to obtain the
marginal pro