| Along with the deepening of Chinese economic reform and the reform of market economic system, all economic regions are seeing a certain degree of improvement in China. However, due to various geographic locations, economic policies, labor and other factors, the economic development status between different regions are different. In the year1978, Shandong and Liaoning has a GDP of almost the same level, but in last year, Shandong has gained a GDP of more than5000billion RMB, while the GDP of Liaoning province is2480billion, about a half of that of Shandong. To study about the differences between these2provinces, in this paper, I selected a series of economic development indicators of the31cities in Shandong and Liaoning provinces, using cluster analysis methods, I comparatively analyzed the regional development differences between the two province. And through the comparation of current economic status, I posed some reasonable improvement suggestions according to the comparation result.In this paper, I chose a dataset in which all the data are related with economics development status of Shandong and Liaoning provinces, and the data range from2002to2011. The main content of this paper is listed in the following paragraphs.Chapter1, introduction, in this section, I mainly introduced the background and purpose of comparing Liaoning with Shandong province, and then gave an introduction of the content and the significance of this study, and then at the end of this chapter, I pointed out the innovations and the shortcomings of this article.Chapter2, the study status of city’s economic development, tell about studies in China and around the world.Chapter3, theory discussion. In the first part, I introduced about the theory of reduction of data dimensionality, in this section, I mainly introduced the processing of panel data. I firstly introduced the theory of principal component analysis and on the basis of the principal component analysis, I realized the reduction of data dimensionality on the time dimension, transform the three-dimensional data into two-dimensional data, which is prepared for subsequent analysis. And in the second part, introductions of cluster analysis theory. The first part is an overview of cluster analysis, including definitions, characteristics, applications cluster analysis, cluster analysis, and typical steps; the second part describes the detailed analysis of the cluster analysis and compared two commonly used methods:hierarchical clustering methods and K-means clustering method.Chapter4, empirical analysis, in this section, I selected SPSS as my analytical tool. Analysis including a total of two steps, the first step is the data dimensionality reduction, in this step, I calculated the scores of each indicator with principal analysis method; then I dealt with the clustering analysis of different indicators in different cities, I created3index systems:overall economic conditions index system, living standard index system and industrial structure index system. For each system, I chose an appropriate clustering method and gain a result of each group.Chapter5, suggestions, in this paper, I also proposed some policy recommendations to improve the economic status of Liaoning Province, such as promoting the development balance among different cities, the enhancement of communications between a rich city and a relatively poor city, increasing the income of workers to improve living standard.The main innovation in this paper lies in indicator selection and the different assessment method, and besides, the processing of the original dataset is also an innovation.The shortcomings of this paper are mainly about the potential errors, firstly, the clustering methods are the most used ones, and their inherent defects are more and more clear, and secondly, the clustering method I chose may not be the best one, and thirdly, the data I chose are mostly suitable, but it is inevitable that some of the best indicators are missed, thus there may be some mistakes, and so on. |
PDF Full Text Request