中文摘要：

为解决Kuznets＂倒U假说＂检验中参数模型设定误差问题,采用非参数方式考察经济增长对不平等的非线性影响,而控制变量采用了局部线性的设定方式,从而建立了一个可用于检验＂倒U假说＂的半参数模型。通过半参数模型,对中国城镇居民和农村居民经济增长与收入不平等之间的＂倒U假说＂进行了再检验,并将检验结果与参数模型进行对比。实证结果表明：城镇居民Gini系数与人均GDP的关系呈近似＂倒U型＂曲线且处于上升的趋势;农村居民Gini系数与人均GDP的关系呈现弱的＂U型＂关系,拒绝了＂倒U假说＂。

英文摘要：

In the traditional parametric model,Inverted-U Hypothesis is usually represented by quadratic form.Nonparametric method is used to inspect the nonlinear impact of economic growth on inequality in order to solve the problem of model specification error.Meanwhile,the method of local linear is used on the setting of controlled variable so as to avoid the problem of curse of dimensionality,and in that way a semi-parametric model that can used to test the Inverted-U Hypothesis is founded.The paper retests whether the Inverted-U Hypothesis exists in the relationship between income inequality and the economic growth of residents in urban and rural areas.And the result of the test is compared with the parametric model.The research shows that the relationship between Gini coefficient of urban residents and GDP per-capita turns to be an approximate Inverted-U curve and the curve tends to rise;the relationship between Gini coefficient of rural residents and GDP per-capita turns to be an approximate linear relation and rejected the inverted-U hypothesis.