基础设施投资最优规模检验与规模预测

发布时间：2018-08-04 20:35

【摘要】：基础设施既作为一项政府公共支出可以直接促进经济增长，同时兼具“溢出效应”和“网络效应”等特殊性质，是一国经济发展必不可少的重要物质基础。基础设施的投资建设对国民经济、行业发展以及所在地区的社会经济产生重大而深远的影响，投资形成的后果呈连锁反应和投资乘数效应。然而其投资规模并非越大越好，过大的基础设施投资规模会对私人资本产生挤出效应，因此存在一个帕累托最优的问题，应与私人物品的供应、总体经济的增长等相协调。鉴于此，如何衡量各省基础设施的实际投资是否达到最优规模？如何根据各项基础设施投资的缺口和紧缺程度，做出合理的投资安排？不同地区之间对基础设施的需求类型是否是相同的？以及如何科学、全面、可靠地对基础设施的投资规模进行优化预测？对于我国基础设施资金规划以及基础设施项目投融资管理、确定政府补贴额度有着重要的理论意义和现实意义。本文以基础设施投资规模及其相关影响因素为研究对象，首先利用DEMATEL因素分析模型识别影响基础设施投资规模的关键因素，其次将最优政府支出模型的政府公共支出分项引入生产函数，得到基础设施投资的边际产出与最优规模理论模型。进而采用中国1996-2012年省级面板数据进行实证研究，探讨我国以及东、中、西部地区基础设施投资的产出效应和最优投资规模，并将我国的实际投资规模和最优规模水平相比较，检验各项基础设施的实际投资是否达到最优规模？以及为实现投资收益最大化，东中西地区各项基础设施的投资次序建议。在此基础上，将基于改进的GA-PSO-V-SVR模型应用于基础设施投资规模预测，，同时为得到有效输入变量，本文结合前文DEMATEL因素分析模型所得结果，以剔除基础设施投资规模影响因素的冗余变量，从而达到降维去噪的目的。最后以全国以及东中西部地区1990—2012年的基础设施投资相关数据进行实例研究，预测全国以及各地区各项基础设施2013年、2015年及2020年的投资规模，为政府基础设施投资提供决策参考。
[Abstract]:As a government public expenditure, infrastructure can directly promote economic growth, and it also has special properties such as "spillover effect" and "network effect". It is an indispensable material basis for a country's economic development. The investment and construction of infrastructure have great and far-reaching influence on the national economy, the development of the industry and the social economy of the region, and the consequences of the investment are chain reaction and investment multiplier effect. However, the larger the investment scale is, the better, and the larger the infrastructure investment scale is, the more the private capital will be squeezed out. Therefore, there is a Pareto optimal problem, which should be coordinated with the supply of private goods and the growth of the overall economy. In view of this, how can we measure whether the actual investment in infrastructure in the provinces is optimal? How to make reasonable investment arrangements according to the gap and shortage of infrastructure investment? Is the type of infrastructure demand the same across regions? And how to optimize and predict the scale of infrastructure investment scientifically, comprehensively and reliably? It is of great theoretical and practical significance to determine the amount of government subsidy for the planning of infrastructure funds and the management of investment and financing of infrastructure projects in China. In this paper, the scale of infrastructure investment and its related influencing factors are studied. Firstly, the key factors affecting the scale of infrastructure investment are identified by using the DEMATEL factor analysis model. Secondly, the government public expenditure of the optimal government expenditure model is introduced into the production function, and the marginal output and optimal scale model of infrastructure investment are obtained. Then using the provincial panel data from 1996 to 2012 in China, the paper discusses the output effect and optimal investment scale of infrastructure investment in China and the east, middle and west regions, and compares the actual investment scale with the optimal scale level in China. Test whether the actual investment in infrastructure is optimal? And in order to maximize investment returns, East, West and East infrastructure investment order suggestions. On this basis, the improved GA-PSO-V-SVR model is applied to the prediction of the scale of infrastructure investment. In order to obtain the effective input variables, this paper combines the previous DEMATEL factor analysis model to obtain the results. In order to eliminate the redundant variables which influence the scale of infrastructure investment, the dimension reduction and denoising can be achieved. Finally, using the relevant data of infrastructure investment from 1990 to 2012 in China and the eastern, western and central regions, the paper forecasts the investment scale of the whole country and all regions in 2013, 2015 and 2020. To provide decision-making reference for government infrastructure investment.
【学位授予单位】：华南理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F299.24

【参考文献】