摘要
减少统计数据误差,提高数据质量对保障政府和企业的合理决策以及大量实证研究结论的可靠性具有重要意义。论文鉴于众多社会经济数据按企业注册地址进行统计,提出造成数据统计误差的新视角——企业实体位置与注册地址的空间分离,并以A股上市公司为例,考察公司数量、公司指标、发展环境指标(地区指标)以及地理距离等4类数据在省级尺度上按注册地址统计的误差幅度。研究显示,若按注册地址统计:各省上市公司数量的误差率平均为14.23%,6项公司指标归并的误差率平均为24.84%,且不同省份、不同指标的误差率差异较大;各行业上市公司按注册地址统计其8项地区指标平均误差率为3.04%,误差率较小;许多公司在总部所在地聘用会计师事务所,造成按注册地址计算上市公司到会计师事务所的地理距离会明显高于实际值。总体而言,企业实体与注册地址的分离现象对各类按后者统计的社会经济数据造成了不可忽视的误差影响,经济地理学对此应予以重视。
Statistical data errors not only mislead the analysis,decision making,and actions of the government,business,and individuals,but also cast doubt on the conclusions of numerous academic studies.Among these data,social-economic data related to businesses is an important target of statistical work,and the determination of business locations is a prerequisite for statistical data on businesses.Currently,the registered address serves as the business location in many social-economic data,but the phenomenon of spatial separation between confirmed business entities and registered addresses may result in significant errors when statistics are conducted based on registered addresses.This article argued that the spatial separation between business entity locations and registered addresses is a new perspective for understanding and studying statistical data errors,providing the possibility to measure actual data errors.In addition to the registered address,listed companies also disclose their office address,which serves as a breakthrough point for quantifying the separation between business entities and registered addresses and studying the statistical errors in data.Taking the office address as the geographical location of A-share listed company headquarters and using various types of data based on this location as the correct values,this study compared and analyzed the errors in data collected based on the registered address(including the number of companies,company indicators,regional indicators,and geographical distances).The study showed that when statistics are generated based on the registered address,the 233 A-share listed companies examined in this study are misclassified in terms of their province(that is,the headquarters are located in a province that is different from the registered address),with most of these companies being located in economically developed regions such as Beijing,Shanghai,and Guangdong,while the distribution of registered addresses is relatively scattered.Influenced by these companies,the average error rate of the number of companies calculated based on the registered address is 14.23%per province,indicating a more imbalanced distribution of the actual headquarter locations of listed companies compared to the registered addresses.The combined average error rate of the six company indicators is 24.84%,showing significant variation in error rates across different provinces and indicators.The average error rate of calculating eight regional development indicators based on the registered address for companies in different industries is 3.04%,indicating a relatively small error rate.The statistical error in calculating the geographical distance between companies and accounting firms based on the registered address depends on the location relationship between the headquarters,registered addresses,and accounting firms.Many companies hire accounting firms in their headquarter locations,leading to a significantly higher geographical distance calculated based on the registered address compared to the actual value.The issue of statistical data errors caused by the spatial separation between business entities and registered addresses will persist in the long term.Given the significant impact of data errors on government decision making,business strategy formulation,academic research,and so on,it is essential for the academic community,especially those in economic geography,to pay attention to the statistical data errors caused by the separation between business entities and registered addresses and their negative effects.
作者
胡国建
陆玉麒
钟业喜
HU Guojian;LU Yuqi;ZHONG Yexi(School of Geography and Environment,Jiangxi Normal University,Nanchang 330022,China;Key Laboratory of Poyang Lake Wetland and Watershed Research,Ministry of Education,Jiangxi Normal University,Nanchang 330022,China;School of Geography Science,Nanjing Normal University,Nanjing 210023,China)
出处
《地理科学进展》
CSSCI
CSCD
北大核心
2023年第12期2309-2323,共15页
Progress in Geography
基金
国家自然科学基金项目(42171171,42361050)。
关键词
企业实体
注册地址
总部
上市公司
数据误差
enterprise entity
registered address
headquarter
listed company
data error
作者简介
第一作者:胡国建(1992-),男,江西九江人,讲师,主要研究方向为经济地理与区域发展。E-mail:guojianhu1992@163.com;通信作者:钟业喜(1973-),男,江西赣州人,教授,研究方向为经济地理与空间规划。E-mail:zhongyexi@jxnu.edu.cn。
