摘要
研究目标:探讨在二元信息基础上引入第三元信息时提升SAM表平衡方法精度的边界条件,考察综合利用流量、系数矩阵信息时如何选择加权平均形式以提升平衡方法的精度。研究方法:基于蒙特卡罗模拟思想,在SAM表中引入均匀分布和正态分布随机误差项,通过均方根误差指标来比较平衡方法的有效性。研究发现:在平衡SAM表时,如果流量信息精度较高,则应采取加权算术平均的形式构建目标函数;如果行系数或列系数矩阵精度较高,则应该采取加权几何平均的形式构建目标函数。在额外信息精度较高的情形下,引入额外信息将有助于提升平衡后SAM表的精度。在额外信息精度较差的情形下,当在列系数和行系数矩阵信息基础上以加权几何平均形式引入精度较差的流量信息时,可提高平衡后SAM表的精度;在其余情形下,额外引入精度较差的信息并不会显著提高平衡后SAM表的精度。研究创新:探讨了在目标函数中引入额外信息时提升平衡方法精度的边界条件,并比较了加权平均形式选择对估计精度的影响。研究价值:本研究对于提升SAM表的精度具有重要参考价值。
Research Objective:This paper discusses the boundary conditions of improving SAM table accuracy after introducing additional information on the basis of binary information,and investigates how to choose the weighted average form to improve the accuracy of the balancing method when using the information of flow rate and coefficient matrix comprehensively.Research Methods:Based on the idea of Monte Carlo simulation,we introduce random error term of uniform and normal distribution into the balanced SAM.Then robustness of the balancing methods are compared using RMSE index.Research Findings:When balancing the SAM table,if accuracy of the flow is higher,the weighted arithmetic average should be adopted to construct the objective function.If the accuracy of the row coefficient or column coefficient matrix is higher,the weighted geometric average should be adopted to construct the objective function.In the case of high accuracy of additional information,the introduction of additional information will help improve the accuracy of the balanced SAM table.On the basis of column coefficient and row coefficient information,accuracy of the estimated SAM can be improved by introducing the flow information with poor accuracy when adopting weighted geometric average.In other cases,if the accuracy of additional information introduced is poor,it will not increase the accuracy of the balanced SAM table.Research Innovations:In this paper,the boundary conditions of improving SAM table accuracy are discussed when additional information is introduced into the objective function of balancing method,and the selection of weighted average form is also discussed.Research Value:This study is useful for improving the accuracy of SAM table.
作者
涂涛涛
马强
Tu Taotao;Ma Qiang(College of Economics and Management,Huazhong Agricultural University)
出处
《数量经济技术经济研究》
CSSCI
CSCD
北大核心
2021年第6期166-180,共15页
Journal of Quantitative & Technological Economics
基金
国家自然科学基金青年基金项目(71503092)
国家自然科学基金国际(地区)合作与交流项目(71461010701)
中央高校基本科研业务费专项资金资助项目(项目批准号2662020JGPY009)的资助。
