Frontiers in Business, Economics and Management ISSN: 2766-824X | Vol. 5, No. 3, 2022 218 Based on the Principal Component Analysis of Our Country's Regional Economic Development Level of Comprehensive Evaluation Yangyang Qian, Yunyun Wang School of Anhui University of Finance and Economics, Bengbu, 233000, China Abstract: China has a vast territory, but due to the influence of historical development, resource endowment, natural conditions, population, policy and other factors, there are great differences in the level of economic development in different regions. In this paper, through the establishment of China's provinces, municipalities, autonomous regions of social and economic development research, the use of principal component analysis method, the level of economic development in China's various regions to make analysis, put forward the corresponding suggestions. Keywords: Economic development, Principal component analysis, Comprehensive evaluation, Cluster analysis. 1. Introduction The economic development of various regions in China is affected and restricted by the natural conditions, resource sharing, resource development, utilization, population quality, economic policy and so on. Regional economic development refers to the process of a region's development from poverty and backwardness to the modernization of economic and social life, which is affected by various factors, the level of economic development in the thirty-one provinces, municipalities and autonomous regions of our country differs greatly, and the level of regional economic development is unbalanced. In response to the problem of unbalanced regional economic development, the state has successively formulated a series of regional development strategies, including: Western Development, revitalizing old industrial bases in northeast China, promoting the rise of central China and supporting the economic opening of eastern China. In order to promote the coordinated development of different regions. In recent years, the state has carried on the elaboration and the adjustment to the regional development strategy. Set up a number of national-level comprehensive reform pilot zones, set different strategic objectives, functions and tasks. With the support of national policy and the continuous efforts of local governments, the economic growth rate of our Western, central and northeastern regions exceeds that of the eastern regions. But the east still has a big advantage over the rest of the country, thanks to its Base and large economic base. Specific to each province, municipalities, autonomous regions there are also obvious differences. In order to comprehensively evaluate the economic development level of every province, municipality and autonomous region in our country, this paper establishes a set of comprehensive index system, and adopts the corresponding index sample data of every province, municipality and autonomous region, using principal component analysis method to calculate the economic development of provinces, municipalities and autonomous regions and the existing differences. 2. The Establishment of Evaluation Index System and The Selection of Evaluation Methods 2.1. The selection of Evaluation Index In accordance with the principle that indicators can objectively, systematically and comprehensively reflect the level of economic development of the region, taking into account specific research issues and taking into account the availability of data, in this paper, the following indexes are selected: x 1-gdp per capita (billion yuan) , x 2-urban population ratio (%) , X 3-tertiary sector of the economy value added in GDP; X 4 -- fiscal revenue as a proportion of GDP (billion yuan) , X 5 -- per capita disposable income (yuan) , x 6 -- per capita consumption expenditure (yuan) , x 7 -- per 100 households (number of computers) , X 5 -- per capita consumption expenditure (yuan) , x 6 -- per capita consumption expenditure (yuan) , x 7 -- per 100 households (number of computers) , X 5 -- per 100 households (number of computers) , X 5 -- per 100 households (number of computers) , X 6 -- per 100 households ( X 8 -- retail sales of consumer goods (billion yuan) . 2.2. Introduction of the Evaluation Method Principal component analysis was put forward by Hotllin in 1933. This method tries to recombine the original indicators into a new set of unrelated composite indicators to replace the original indicators, and at the same time, according to the actual needs, a few fewer composite indicators can be obtained from them to reflect the information of the original indicators as much as possible. The new index variables are small in number and independent of each other, keeping the main information of the original index variables. This method has many advantages: first, it eliminates the correlation between the samples of evaluation indicators; second, the extracted principal components keep the main information of the original indicators, reducing the workload; third, the analysis process objectively generates the weight of indicators, which can distinguish the role of each indicator in the comprehensive evaluation. It avoids the influence of subjective factors. 219 3. Principal Component Analysis Process Data were first normalized with Stata software, followed by a KMO test with the normalized data to probe for correlations between variables, as shown in Figure, where the value of the KMO statistic equals 0.705 & LT; 1, the fitting effect is good and principal component analysis can be used. Figure 1. KMO inspection Using Stata software, we can get the correlation coefficient matrix table of the index sample, such as Figure 2, the eigenvalues and eigenvectors of the correlation coefficient matrix, the contribution rate of each eigenvalue, and finally get the variance contribution analysis table, figure 3: Figure 2. Correlation coefficient matrix table Figure 3. Contribution to variance From the graph above, we can see from Table 1 that the cumulative contribution rate of the former eigenvalues has reached 77.58% , which indicates that the first three factors can reflect more than 77.58% of all information, the eigenvalue of the third component is less than 1, so take the first two factors as the principal component to calculate the principal component score coefficient matrix, and calculate the principal component coefficients of each index according to the score coefficient data matrix. Table 1. Principal Component Score Variable Comp1 Comp2 X1 0.4418 0.1457 X2 0.4304 0.0633 X3 0.2765 -0.5153 X4 0.2759 -0.5712 X5 0.4646 0.0238 X6 0.1096 0.2610 X7 0.4480 0.1389 X8 0.1962 0.5431 As shown in Figure 4, the greater the absolute value of the load coefficient, the greater the effect on the principal component and the stronger the interpretation of the principal component. The first principal component has a nearly equal positive load on all variables, which can be called the level of comprehensive economic development, while the second principal component has a larger positive load on per capita consumption expenditure (x 6) and retail sales of social consumer goods (x 8), it reflects the economic development orientation mainly affected by the resident consumption, which can be called the level of consumption-oriented economic development. KMO = 0.705 Kaiser-Meyer-Olkin Measure of Sampling Adequacy H0: variables are not intercorrelated p-value = 0.000 Degrees of freedom = 28 Chi-square = 239.741 Bartlett test of sphericity Det = 0.000 Determinant of the correlation matrix zx8 0.4927 0.3426 -0.1175 -0.2707 0.3916 0.1372 0.4870 1.0000 zx7 0.8712 0.8853 0.3837 0.3974 0.9325 0.2403 1.0000 zx6 0.2426 0.1810 -0.0169 -0.0615 0.1681 1.0000 zx5 0.9490 0.8744 0.5106 0.5220 1.0000 zx4 0.3781 0.4365 0.8354 1.0000 zx3 0.3826 0.3833 1.0000 zx2 0.8156 1.0000 zx1 1.0000 zx1 zx2 zx3 zx4 zx5 zx6 zx7 zx8 Comp8 .0162312 . 0.0020 1.0000 Comp7 .0938123 .0775811 0.0117 0.9980 Comp6 .128163 .0343506 0.0160 0.9862 Comp5 .182758 .0545955 0.0228 0.9702 Comp4 .459809 .27705 0.0575 0.9474 Comp3 .912608 .4528 0.1141 0.8899 Comp2 1.77598 .863368 0.2220 0.7758 Comp1 4.43064 2.65467 0.5538 0.5538 Component Eigenvalue Difference Proportion Cumulative 220 Figure 4. Principal component interpretation The principal component expression is obtained from the coefficient matrix: F1=0.4418x1+0.4304x2+0.2765x3+0.2759x4+0.4646x5+ 0.1096x6+0.4480x7+0.962x8 F2=0.1457x1+0.0633x2-0.5153x3- 0.5712x4+0.0238x5+0.2610x6+0.1389x7+0.5431x8 From the calculation of the score and the final ranking can be seen that our 31 provinces, municipalities, autonomous regions of the level of economic development there are great differences. The provinces and municipalities directly under the central government in the eastern region all have high scores and the top ranking. Shanghai and Beijing are undoubtedly the two most developed cities in the overall level of economic development, ranking No. 1 and No. 2 respectively, it is worth mentioning that the consumption- oriented economic development components of these two cities ranked lower, at 26th and 28th respectively, so there is still potential for development to further enhance the economic development of developed cities, creating a world- class city could spur consumer spending in Beijing and Shanghai. And far ahead of other cities, while the western provinces, municipalities, and autonomous regions all scored lower, especially Yunnan, Guizhou, Gansu, and Tibet, the first principal component score and the second principal component score are both negative, ranking low, the economic development situation is not optimistic. The cities with the highest scores of the second principal component were Hubei, Fujian, Jiangsu and Guangdong, where consumption had a significant impact on economic development, especially Hubei, which ranked first in the second principal component score, the first principal component is the 27th, so we should improve the comprehensive development level of Hubei, increase basic investment and construction. The principal component scores of Zhejiang, Guangdong, Jiangsu and Fujian are all positive, and their development prospects are good. This result accords with the level of economic development of our country and reflects the unbalanced development of regional economy of our country. In order to analyze the difference of regional economic development level in the whole country more intuitively and accurately, using Stata software to cluster the two principal component scores of 31 provinces, autonomous regions and municipalities directly under the central government, there are four categories of cities with different levels of economic development. Class 1: Shanghai, Beijing. The level of comprehensive economic development is far ahead, but the level of consumption-oriented economic development is not significant; the second category: Hubei. The third category is: Tianjin, Zhejiang, Guangdong, Jiangsu, Fujian, Shandong. The total economic development level of these cities is at a higher level in the country, and the composition of consumption guide is more balanced. The fourth category: Liaoning, Inner Mongolia, Xinjiang, Qinghai, Gansu and so on. These cities have low scores of both principal components and backward economic development. 4. Policy Recommendations In view of the unbalanced development of regional economy and the main problems existing in the process of regional economic development, this paper puts forward the following suggestions: first, we will continue to promote the development of the west, revitalize the old industrial bases in the northeast, promote the rise of the central region, and support the economic opening-up of the eastern region. According to the characteristics of economic development in different stages, the regional economic development strategy is refined and adjusted so as to promote the economic development of backward areas more effectively. Second, according to the comparative advantages of different regions, different policy preferences and technical support should be given. Guide different regions to develop advantageous and characteristic industries. Accelerate the economic development of backward areas. The relevant government departments should perfect the supporting policies and strengthen the policy support so as to attract more investment for the central and western regions. Third, increase investment and promote infrastructure construction in the zx1 zx2 zx3 zx4 zx5 zx6 zx7 zx8 -. 5 0 .5 C o m p o n e n t 2 .1 .2 .3 .4 .5 Component 1 Component loadings 221 western region. The lack of economic growth in the western region leads to the lack of sufficient financial resources, the inability to provide good and convenient infrastructure, and the difficulty in improving the investment environment, which restricts economic growth. Therefore, the central and local governments should increase investment in infrastructure construction in the western region, so that the western region can greatly improve its health, transportation, energy and Water Conservancy, airports, oil and gas pipelines, and so on, improve the investment climate to attract more investment and better achieve industrial upgrading. References [1] Cui Qinyu,Huang Yiting,Yang Guang,Chen Yu. Measuring Green Exposure Levels in Communities of Different Economic Levels at Different Completion Periods: Through the Lens of Social Equity[J]. International Journal of Environmental Research and Public Health,2022,19(15). [2] Wang Yizhen,Sun Ken,Li Li,Lei Yalin,Wu Sanmang,Jiang Yong,Mi Yifeng,Yang Jingjing. The impacts of economic level and air pollution on public health at the micro and macro level[J]. Journal of Cleaner Production,2022,366. [3] Azevedo Jefferson Moraes,Reis Matheus de Paula,Gous Robert M.,Dorigam Juliano Cesar de Paula,Lizana Rony Riveros,Sakomura Nilva Kazue. Response of broilers to dietary balanced protein. 2. Determining the optimum economic level of protein[J]. ANIMAL PRODUCTION SCIENCE,2021,61(14). [4] Chen Jialong,Yang Liuna,Qian Zhenzhu,Sun Mingwei,Yu Honglin,Ma Xiaolei,Wan Chonghua,Yang Yunbin. Cluster analysis of differences in medical economic burden among residents of different economic levels in Guangdong Province, China.[J]. BMC health services research,2020,20(1). [5] Song Yanwu,Liu Hongwei. Internet development, economic level, and port total factor productivity: an empirical study of Yangtze River ports[J]. International Journal of Logistics Research and Applications,2020,23(4). [6] Mustafa Soylak,Recep Saraymen,İbrahim Narin,Mehmet Doğan. Serum zinc levels of a region that has low socio- economic level[J]. Eur J Ther,2019,8(1). [7] Water Resource Planning; Findings from Qassim University Provides New Data about Water Resource Planning (Framework To Establish Economic Level of Leakage for Intermittent Water Supplies In Arid Environments)[J]. Ecology Environment & Conservation,2019. [8] [Ye Duan,Zenglin Han,Hailin Mu,Jun Yang,Yonghua Li. Research on the Impact of Various Emission Reduction Policies on China’s Iron and Steel Industry Production and Economic Level under the Carbon Trading Mechanism[J]. Energies,2019,12(9). [9] Shuai Wang, Zhizhang Li, Chen Wen. The Impact of Credit Support on Low Carbon Economy: An Empirical Study Base on Macro and Micro Economic Level in China[J]. International Journal of Economics, Finance and Management Sciences,2018,6(6).