多阶模型在地区消费差异研究中的应用

 当组间差异显著时，应该运用多阶模型纳入更高阶的场景变量进行分析。对2007年广东省11市7县、区的1600个家庭的消费数据研究表明，不同城市之间居民的消费水平、边际消费倾向都存在显著差异。文章引入微观层面的家庭人均可支配收入（PDI）和是否拥有住房（HOUSE）虚拟变量来解释微观层面的家庭消费差异；引入宏观层面的地区人均可支配收入（GPDI）和是否为发达地区（D）虚拟变量作为场景变量解释家庭消费跨地区的差异。实证结果表明，微观层面变量对家庭消费差异（组内差异）的解释程度达48％以上，两个场景变量对城镇居民消费跨地区差异（组间差异）的解释程度达99％以上。

Multilevel Models' Applications in the Study of the Regional Differences in Consumption

When the sample’s between-group variation is statistically notable,we should use multilevel models and introduce contextual variables.This paper explores the 1600 households’ consumer data,including 11 cities,7 countries or towns in Guangdong province.The conclusions are that notable difference exists in consumption level,marginal propensity to consume among different cities,countries and towns.This paper introduces per capita disposable income of a family(PDI) and dummy variable whether having a house(HOUSE) in micro-lay to explain the consumption difference among households,and introduces the status of development(D) and dummy variable Area per capita disposable income(GPDI) as contextual variables to explain the consumption difference among different cities and countries.Empirical study shows that variables in micro-lay can explain more than 48 percent about intra-class consume difference and the two contextual variables can explain more than 99 percent about the between-group difference.