中国省际共同富裕水平评价研究

Research on the Evaluation of China’s Inter-provincial Common Prosperity Level

共同富裕是社会主义的本质和社会主义发展的最终目标,党的二十大报告强调,要实现“全体人民共同富裕的现代化”。研究共同富裕的相关问题,对于走好共同富裕道路、全面建设社会主义现代化国家、实现共同富裕的理想目标有着理论性的指导作用。当前学者对于共同富裕问题的研究大多集中在对相关内涵概念的辨析,而对于如何测定中国共同富裕的发展过程、各个省份共同富裕的差异性和共同富裕程度分类预测的研究相对较少。文章旨在通过对中国各个省份共同富裕水平进行评价与比较,将东西部地区、南北区域的共同富裕发展状况形象地展现出来,以期为政府部门相关工作提供具有参考性的建议。研究数据来源于《中国统计年鉴》2013—2020年31个省份的相关数据,从经济发展、社会发展、收入消费、文化发展及生态环境五个方面,选取了14个评价指标,以主成分分析法(PCA)进行打分,采用K均值聚类对打分结果进行等级划分,并通过卷积神经网络(CNN)模型对各省份共同富裕水平进行预测与仿真。研究结果表明:(1)从各个等级间评价指标的差异性来看,影响不同省份共同富裕发展状况的指标,主要分布在经济发展、社会发展、收入消费,文化发展四个层面。(2)从发展历程来看,2013—2020年中国社会共同富裕整体呈现积极变化态势,其中2020年发展程度最好。(3)从地区分布来看,东部地区如北京、上海等地经济发展较好,共同富裕程度相比其他地区较好。部分西部地区,人民生活水平不高,实体经济发展仍需完善。(4)从模型准确率来看,CNN模型具有较好的预测效果,平均预测准确率达到91.6%,实验效果优于选取的四种对比模型(MLP、BP、SVM、KNN)。本研究对于均衡发展生产力、消除两极分化具有一定的现实意义。针对研究结果,文章也提出了一些建议,希望对落实共同富裕这一目标,积极推动经济高质量发展具有良好借鉴意义。

Common prosperity is the essence of socialism and the ultimate goal of socialist development.Studying the related issues of common prosperity has a theoretical guiding role for taking the road of common prosperity,building a socialist modernized country in an all-round way,and realizing the ideal goal of common prosperity.At present,scholars’research on common prosperity mostly focuses on the discrimination of related connotation concepts,but there are relatively few studies on how to measure the development process of common prosperity in China,the differences of common prosperity in different regions and the classification prediction of common prosperity degree.The purpose here is to show the image of the common prosperity of the eastern and western regions,the north and south regions by evaluating and comparing the common prosperity levels of various regions in China,so as to provide reference suggestions for the relevant work of government departments.The statistics spanning eight years are selected so that the process of promoting common prosperity by the Party Central Committee can be evaluated in the form of big data.The research data are obtained from the relevant data of 31 provinces and municipalities directly under the central government during the period of 2013-2020 in China Statistical Yearbook.14 evaluation indicators are selected from five aspects:economic development,social development,income consumption,cultural development and ecological environment,scored by principal component analysis(PCA),K-means clustering is used to rank the scoring results,and convolutional neural network(CNN)model is used to predict and simulate the common prosperity level of each region.The research results show that:From the perspective of the differences of evaluation indicators among different levels,the indicators that affect the development of common prosperity in different regions are mainly distributed in four levels:economic development,social development,income consumption,and cultural development;From the perspective of the development process,the overall prosperity of China’s society from 2013 to 2020 showed a positive trend of change,of which 2020 had the best development;From the perspective of regional distribution,the economic development of eastern regions such as Beijing and Shanghai is relatively good.Common prosperity is better than other regions.In some western regions,people’s living standards are not high,and the development of the real economy still needs to be improved;From the perspective of model accuracy,the CNN model has a good prediction effect,with an average prediction accuracy rate of 91.6%,and experimental results are better than the selected four comparison models(MLP,BP,SVM,KNN).Based on the research results,some suggestions are put forward,which are hoped to have good reference significance for implementing the goal of common prosperity and actively promoting high-quality economic development.With the development of the new era,the spiritual connotation of the idea of common prosperity will be enriched continuously,therefore,the evaluation of the level of common prosperity is also a continuous exploration process,and the evaluation methods as well as evaluation indexes will certainly become more comprehensive,accurate and scientific with the depth of research.