Research On The Factors Affecting Income Distribution From The Perspective Of Multi-dimensional Income

This paper establishes a Hierarchical Partial Least-Square(H-PLS)Regression Model from the perspective of multi-dimensional income for residents' income structure on the factors of GDP structure,employment structure,population structure,urbanization level and fiscal expenditure,which provides referential suggestions to discover a new path for income growth,and help the government to make reasonable policies on income distribution through investigating the role of various factors above for the income structure.Residents' income structure data that is made up of wage income,operating income,property income and metastatic income(we call them “sub-income”)is compositional data with “sum to unit”,so are GDP structure,employment structure and population structure.According to the dual economic structure of our nation,this paper establishes two H-PLS regression models for urban residents and rural residents based on the covariance structure of composition data to discuss the systemic impact of independent variables selected in the model on the income structure.According to the results of the two models,the impacts of independent variables on the income structure is different between urban and rural,and also between different incomes included in the income structure.The difference may counteract the efficiency of a policy aimed to improve a sub-income or total income,which enlightens us to be cautious and attach priority to principal contradiction when implement policies.As a result of the conclusive relationship between sub-income and total income,the study on the income structure that based on the multidimensional perspective is valuable and significant for maintaining sustainable growth of income,narrowing the income gap and achieving the ultimate goal of common prosperity.