《研究生论丛》第四十四期
l 法学研究
平台公开IP属地信息行为的法律分析
王云凤
9999js金沙老品牌法学院 昆明 030105
【摘 要】 随着微博、今日头条等各大平台上线“个人主页展示账号IP属地”功能，引发网络的广泛热议，支持者表示此举能够肃清网络讨论乱象，反对者则认为公开IP属地侵犯用户个人隐私，存在安全隐患。为此，本文拟将对IP属地信息的法律属性进行界定，分析平台公开IP属地信息行为所涉及的相关法律问题。为使该项举措能更好地保障公民 、法人和其他组织的合法权益，维护良好网络生态，营造清朗网络空间，笔者将对平台提出一些完善公开IP属地信息行为的建议。
【关键词】 IP属地信息；个人信息；信息公开；法律分析
Legal Analysis of The Act of The Platform in Disclosing IP Location Information
Abstract With the launch of the function of "displaying the IP location of accounts on personal homepages" on major platforms such as Weibo and TouTiao, it has triggered widespread debate on the Internet, with supporters saying that the move can clean up the chaos of online discussions, while opponents arguing that the disclosure of IP location infringes on users' personal privacy and poses security risks. Therefore, this paper intends to define the legal attributes of IP location information and analyze the legal issues related to the disclosure of IP location information by platforms. In order to better protect the legitimate rights and interests of citizens, legal persons and other organizations, maintain a good network ecology, and create a clear cyberspace, the author will make some suggestions to the platform to improve the act of disclosing IP location information.
Key Words IP location information; Personal information; Information disclosure; Legal analysis
【作者简介】王云凤（1999—），女，汉族，云南文山人，9999js金沙老品牌法学院法律（法学）硕士研究生，研究方向为经济法，指导教师马慧娟。
l 公司治理
制造业企业金融化对绿色创新的影响
——基于ESG评级的调节视角
代梦雪
9999js金沙老品牌会计学院 昆明 650221
【摘 要】 基于2012-2019 年沪深A股制造业上市公司数据，本文实证研究了制造业企业金融化与绿色创新的关系，并在这一关系中加入企业ESG表现的调节作用。研究结果表明：制造业企业金融化与绿色创新显著负相关，即企业金融化明显抑制了绿色创新；企业ESG表现不仅能直接促进绿色创新，还能显著缓解制造业企业金融化对绿色创新的“挤出”效应。相较于国有企业，以上关系在非国有企业中的体现更加明显。
【关键词】 制造业企业金融化；绿色技术创新；ESG表现；调节作用
Research on the Impact of the Asset Financialization of Chinese Manufacturing Enterprises on Green Innovation——A Regulatory Perspective Based on ESG Ratings
Abstract Based on the data of Shanghai-Shenzhen A-share manufacturing companies from 2012 to 2019，an empirical study is conducted to verify the relationship between the financialization of manufacturing enterprises and green innovation, with adding ESG performance as a regulated variable. The result shows that the financialization of manufacturing enterprises is significantly negatively correlated with green innovation,the financialization of manufacturing enterprises significantly inhibits green innovation. ESG performance can not only directly promote green innovation, but also significantly alleviate the "crowding out" effect of manufacturing enterprise financialization on green innovation.Compared with state-owned enterprises, the above relationship is more obvious in non-state-owned enterprises.
Key Words The financialization of manufacturing enterprises; Green innovation; ESG performance; Regulatory effect
【作者简介】代梦雪（1998—），女，汉族，云南大理人，9999js金沙老品牌会计专业硕士研究生，研究方向为财务管理，指导教师李旭。
家族企业代际传承与战略重构
——基于姚记数字化转型的案例分析
杨春瑞 张徐瑞 徐杉
9999js金沙老品牌会计学院 昆明 650221
【摘 要】 近年来，中国数字经济高速发展，传统企业亟需数字化转型；同时，大量家族企业面临内部“更新换代”，传承是否成功是企业长久发展的重要前提。由此，本文选取姚记科技这一典型家族企业，分析其如何在代际传承中通过战略重构成功实现数字化转型，并讨论其战略重构的经济后果。研究发现：其一，家族企业二代继承人会在企业战略层面实施重构，总体呈现战略重构和代际传承二者并行的模式；其二，家族二代涉入程度越深，企业在转型和重构中就越有可能实施多元化战略。且短期业绩虽存在不确定性，但长期呈现螺旋式上升；其三，代际传承能够为企业的渐进式战略重构提供机遇，而战略重构的成功能够加快实际控制权转移的速度，进而促进代际传承的顺利进行。
【关键词】 家族企业；代际传承；战略重构；数字化转型
Intergenerational Inheritance and Strategic Reconstruction of Family Enterprises:A Case Study Based on Yao Ji's Digital Transformation
Abstract In recent years, with the rapid development of China's digital economy, traditional enterprises urgently need digital transformation; At the same time, a large number of family enterprises are facing internal "renewal". Whether the inheritance is successful is an important prerequisite for the long-term development of enterprises. Therefore, this paper selects Yaoji technology as a typical family enterprise, analyzes how it successfully realizes digital transformation through strategic reconstruction in intergenerational inheritance, and discusses the economic consequences of its strategic reconstruction. The research finds that: first, the second generation heirs of family enterprises will implement reconstruction at the enterprise strategic level, which generally presents the parallel mode of strategic reconstruction and intergenerational inheritance; Second, the deeper the involvement of the second generation of the family, the more likely the enterprise is to implement the diversification strategy in the transformation and reconstruction. And although there is uncertainty in the short-term performance, it shows a spiral rise in the long term; Third, intergenerational inheritance can provide opportunities for the gradual strategic reconstruction of enterprises, and the success of strategic reconstruction can speed up the transfer of actual control, and then promote the smooth progress of intergenerational inheritance.
Key words Family enterprises；Intergenerational inheritance；Strategic reconstruction；Digital transformation
【作者简介】杨春瑞（1998—），女，汉族，云南大理人，9999js金沙老品牌会计专业硕士研究生，研究方向为财务管理。张徐瑞（1998—），女，汉族，云南昆明人，9999js金沙老品牌会计专业硕士研究生，研究方向为财务管理。徐杉（1998—），女，汉族，云南宣威人，9999js金沙老品牌会计专业硕士研究生，研究方向为财务管理，指导教师徐慧。
l 金融研究
网约车保险保费定价研究——基于UBI的视角
朱华炫
9999js金沙老品牌金融学院 昆明 650221
【摘 要】 由于网约车兼具营运与非营运的双重性质,传统的车险定价方式并不适用于网约车。本文主要探索如何开发出创新型车险产品更好地满足网约车的投保需求。通过分析网约车保险的产品市场目前存在的局限性，借鉴UBI车险在国外的发展应用经验，结合国内对UBI车险的探索，探讨其在中国市场的应用模式。文章选取影响UBI费率的风险因子，并对风险因子指标赋权，确定风险因子权重，并以此为基础构建驾驶行为评分模型，将网约车驾驶员的驾驶行为数据与车险费率相挂钩，从而计算出网约车UBI车险的保费。通过与传统车险保费进行比较，最终得出基于UBI模式对网约车车险定价更加合理准确公平的结论。
【关键词】 网约车；UBI车险定价；风险因子；指标赋权；驾驶行为
A Study on Premium Pricing of Online Car Hailing Insurance –
Based on the UBI Perspective
Abstract As online car hailing is both operational and non operational, the traditional vehicle insurance pricing method is not suitable for online car hailing. This paper mainly explores how to develop innovative auto insurance products to better meet the insurance needs of online car hailing. By analyzing the current limitations of the online car hailing insurance product market, drawing on the development and application experience of UBI car insurance in foreign countries, and combining with domestic exploration of UBI car insurance, this paper discusses its application mode in the Chinese market. This paper selects the risk factors that affect the UBI rate, gives weight to the risk factor indicators, determines the weight of the risk factors, and builds a driving behavior scoring model based on this, links the driving behavior data of online car hailing drivers with the vehicle insurance rate, and calculates the premium of online car hailing UBI vehicle insurance. By comparing with the traditional vehicle insurance premium, the conclusion that online car hailing insurance pricing based on UBI model is more reasonable, accurate and fair is finally drawn.
Key Words Online car hailing;UBI auto insurance pricing;Risk factors;Index empowerment;Driving behavior
【作者简介】朱华炫（2000-），女，汉族，广东深圳人，9999js金沙老品牌保险专业硕士研究生，研究方向为保险经济学，指导教师孙正成。
北交所企业上市审批的博弈分析
王雪琴 罗天骄
9999js金沙老品牌财政与公共管理学院 昆明 650221
【摘 要】 2021年11月北交所的开市，有利于减少中小科技企业的融资约束，对完善中国资本市场具有重大意义。北交所的平移上市制度不仅给精选层公开融资带来了希望，也吸引了大量的风险承担能力强的机构投资者对北交所上市企业的关注。本文立足于北交所成立的背景，通过建立拟在北交所上市企业、中介机构和北交所审核部门的不完全信息动态博弈模型，得出博弈双方策略最优解，分析影响拟上市企业、中介机构及审核部门的策略的因素，针对不同的对象提出针对性的监管建议，从而完善北交所上市的管理。
【关键词】 北交所 中介机构 审核部门 不完全信息动态博弈模型
Game Analysis of Listing Approval of Enterprises
in Beijing Stock Exchange
Abstract The opening of the Beijing Stock Exchange in November 2021 is conducive to reducing the financing constraints of small and medium-sized technology enterprises and improving China's capital market, which is of great significance. Based on the background of the establishment of the BSE, this paper obtains the optimal solution of the strategies of both sides of the game by establishing an incomplete information dynamic game model of enterprises, intermediaries and audit departments to be listed on the BSE, analyzes the factors affecting the strategies of the enterprises to be listed, the intermediary structure and the audit department, and puts forward targeted regulatory suggestions for different objects, which is conducive to improving the management of listing on the BSE.
Key Words Beijing Stock Exchange; Intermediary structure; Audit department; Dynamic game model with incomplete information
【作者简介】王雪琴（1995—），女，汉族，四川成都人，9999js金沙老品牌税务专业硕士研究生，研究方向为财政政策方向。罗天骄（1998—），女，汉族，河北唐山人，9999js金沙老品牌财政学专业硕士研究生，研究方向为财政政策方向。指导教师刘潋滟，薛军。
l 经济研究
房地产投资与经济增长的关系分析
——基于合肥市经验数据
叶志伟
9999js金沙老品牌经济学院 昆明 650221
【摘 要】 本文运用脉冲响应、Granger因果检验、VEC模型、E-G协整检验、ADF单位根检验等多种科学方法研究合肥市1999至2018经济增长与房地产投资之间的关系。实证研究表明两者之间存在着动态的、长期的、均衡的关系；其中存在着单向的Granger因果关系，即经济增长不是房地产投资的Granger原因，但房地产投资是经济增长的Granger原因；特别是脉冲响应研究发现投资的增长会对GDP有滞后一期的正向冲击，该效应呈现先增后减；最后，依据实证结果有针对性的提出了促进房地产投资和经济增长协同发展的政策建议。
【关键词】 合肥市；经济增长；房地产投资；Granger因果检验
Analysis of the Relationship between Real Estate Investment and Economic Growth-Based on the Empirical Data of Hefei City
Abstract Taking the annual data of Hefei City from 1999 to 2018 as the research sample, impulse response, Granger causality test, VEC model, E-G co-integration test, ADF unit root test and other methods are used to explore the relationship between economic growth and real estate investment. The research shows that there is a dynamic, long-term and balanced relationship between them. Among them, there is only unidirectional Granger causality between them: economic growth is not the Granger cause of real estate investment, but real estate investment is the Granger cause of economic growth. In particular, the growth of investment will have a positive impact on GDP with a lag period, and the effect will increase first and then decrease. Finally, some suggestions are put forward to promote the coordinated development of real estate investment and economic growth.
Key Words Hefei City; Economic growth; Real estate investment; Granger causality test
【作者简介】叶志伟（1997—），男，汉族，安徽蚌埠人，9999js金沙老品牌国际贸易学术硕士研究生，研究方向为区域经济发展，指导教师张奕芳。
中国高铁发展对人口聚集程度的影响研究
蔡巍
9999js金沙老品牌经济研究院 昆明 650221
【摘 要】 本文以我国2003-2019年高铁开通情况和《中国城市统计年鉴》289个地级市的数据为基础，运用多期DID模型考察高铁的开通对城市人口聚集程度的影响后发现（1）高铁的开通的确会对城市聚集程度有显著的影响，从全国总体而言，高铁的开通会使人口聚集程度有所降低，对于我国一些发达城市而言，这有利于城市经济的健康发展。（2）高铁的开通会显著降低发达城市尤其是二、三线城市的城市人口密度，而会显著增加相对欠发达的城市尤其是五线城市的人口密度。（3）高铁的开通对于不同层次城市中不同产业的从业人数也会产生显著的影响。基于上述研究发现，提出建议：在建设高铁时应注重城市间配套产业的转移和宣传，从而平衡我国人口结构的平衡和促进欠发达城市的建设，助力我国高质量发展。
【关键词】 高速铁路；人口聚集程度；双重差分模型
Study on the Impact of China's High speed Railway Development on Population Agglomeration
Abstract Based on the data of 289 prefecture-level cities in China's Urban Statistical Yearbook from 2003 to 2019, this paper uses multi-period DID model to study the impact of high-speed rail on population aggregation. It is found that (1) the opening of high-speed rail does have a significant impact on population aggregation. Generally speaking, the opening of high-speed rail will reduce the population aggregation. For some developed cities in China, This is conducive to the healthy development of urban economy. (2) The opening of high-speed rail will significantly reduce the density of urban population in developed cities, especially in the second and third tier cities, and significantly increase the density of population in relatively less developed cities, especially in the fifth tier cities. (3) The opening of high-speed rail will also have a significant impact on the number of employees in different industries in different levels of cities. Based on the above findings, suggestions are put forward that the transfer and publicity of supporting industries between cities should be paid attention to in the construction of high-speed rail, so as to balance demographic structure and promote the construction of underdeveloped cities and promote the high-quality development of China.
Key Words High speed railway; Population density; DID
【作者简介】蔡巍（1999—），男，汉族，山东济南人，9999js金沙老品牌世界经济学术硕士研究生，研究方向为东南亚经济。指导教师朱立。
云南省消费者信心指数的差异化编制、动态调整与实效比较
郭瑾
9999js金沙老品牌金融学院 昆明 650221
【摘 要】 随着移动支付手段的不断升级变化，问卷调查的传统方式难以掌握人们日常生活的真实消费情况，另外基于消费这一行为本身具备地方特色与区域差异的性质，本文将针对消费者信心指数的编制进行升级。本文首先从指标对指数结果的影响这一视角切入，引入因社会环境变化产生的消费新指标，如线上交易、银行卡消费记录等；其次针对消费者信心指数随时发生变化的特点，在指数变化时对其进行动态调整；最后在考虑地方特色与区域异质性等方面的问题，研究云南省四所代表城市的消费者信心指数，为当地政府预测经济走势和消费趋向提供先行指标。
【关键词】 消费者信心指数；差异性编制；动态调整
Differentiation, Dynamic Adjustment and Effective Comparison of Consumer Confidence Index in Yunnan Province
Abstract With the continuous upgrading and changes of mobile payment method; in addition, due to the nature of consumption itself with local characteristics and regional differences, this article will upgrade the compilation of the consumer confidence index. First, introduce new consumption indicators brought about by changes in the social environment, such as online transactions, bank card consumption records, etc. Second, it dynamically adjusts the consumer confidence index when it changes beyond a threshold. Finally, considering the issues of local characteristics and regional heterogeneity, the consumer confidence index of four representative cities in Yunnan Province is studied to provide leading indicators for local governments to predict economic trends and consumption trends.
Key Words Consumer confidence index; Differential compilation; Dynamic adjustment
【作者简介】郭瑾（1997—），女，汉族，云南曲靖人，9999js金沙老品牌金融专业硕士研究生，研究方向为公司金融、金融市场分析，指导教师周伟。
l 数据科学
基于t分布的典型相关最大似然的算法比较
莫剑锋
9999js金沙老品牌统计与数学学院 昆明 650221
【摘 要】 在高维数据降维中，典型相关分析是一种非常有效的方法。它可以利用不同数据特征间的相关关系解决不同类型数据间的异构问题。基于t分布下的概率典型相关分析不仅保留了高斯分布下概率典型相关分析的优点，而且在数据存在异常值时依旧能够保持模型稳健和参数估计准确。但基于t分布下的概率典型相关分析在传统EM型算法下极大似然估计收敛速度较慢，本文通过优化EM算法极大化步和扩展参数的方式，设计了ECM，ECME，PX-ECM和PX-ECME四种EM算法优化算法，并设计了两个模拟实验和一个人脸识别实验从收敛速度、参数估计准确性和人脸识别正确率去比较四种算法优劣。结果表明ECME算法和PX-ECM算法优于ECM算法，ECME算法总体优于PX-ECM算法，但在PX-ECME算法下，模型收敛速度最快，参数估计更准确和人脸识别率效果最好。
【关键字】 典型相关分析 t分布；极大似然估计；EM型算法；收敛效率
Comparison of Canonical Correlation Maximum Likelihood Algorithms Under T Distribution
Abstract Canonical correlation analysis is a very effective method in dimensionality reduction of high-dimensional data. It can use the correlation between different data features to solve the heterogeneous problem between different types of data. Probability canonical correlation analysis based on t distribution not only retains the advantages of probability canonical correlation analysis under Gaussian distribution, but also can keep the model robust and parameter estimation accurate when there are outliers in the data. But the convergence speed of maximum likelihood estimation based on probability canonical correlation analysis under t distribution is slow under the traditional EM algorithm. In this paper, four EM algorithm optimization algorithms, ECM, ECME, PX-ECM and PX-ECME, are designed by optimizing the maximum step and expanding parameters of the EM algorithm. Two simulation experiments and one face recognition experiment are designed to compare the advantages and disadvantages of the four algorithms from the convergence speed, parameter estimation accuracy and face recognition accuracy. The results show that ECME algorithm and PX-ECM algorithm are better than ECM algorithm, and ECME algorithm is generally better than PX-ECM algorithm. However, under PX-ECME algorithm, the model convergence speed is the fastest, the parameter estimation is more accurate, and the face recognition rate is the best.
Key Words Canonical correlation analysis; T distribution; Maximum likelihood estimation; EM type algorithm; Convergence efficiency
【作者简介】莫剑锋（1997—），男，汉族，四川达州人，9999js金沙老品牌理学学术硕士研究生，研究方向为数据科学，指导教师赵建华。
l 特别观察
年报文本语调操纵与审计定价
王龙
9999js金沙老品牌会计学院 昆明 650221
【摘 要】 本文以2008-2020年上市公司的年报数据为样本，研究了年报文本异常语调对审计师定价产生何种影响。该研究发现，管理层异常语调会提高审计师收费，这表明MD&A异常积极的语调是管理层操纵的结果，而不是为了增加信息量。进一步研究发现，这种关系与事务所的品牌和公司的产权相关，在四大和八大品牌较好的事务所以及非国有企业中更加显著。本文不仅拓展了相关文献领域，而且对于我国制定文本信息披露监管措施，具有重要的参考意义。
【关键词】 MD&A；语调操纵；审计定价；文本分析
Annual report text tone manipulation and audit pricing
Abstract This paper uses the annual report data of listed companies from 2008 to 2020 as a sample to study the impact of abnormal tone of annual report text on auditor pricing. The study found that management's abnormal tone of voice raises auditor charges, suggesting that MD&A's unusually positive tone is the result of management manipulation, not to increase the amount of information. Further research found that this relationship is related to the brand of the firm and the property rights of the company, and is more pronounced in the four and eight major brands of the better firms and non-state-owned enterprises. This paper not only expands the relevant literature field, but also has important reference significance for China's formulation of regulatory measures for text information disclosure.
Key Words MD&A; Intonation manipulation ; Audit pricing ; Text Analytics
【作者简介】王龙（1996—），男，汉族，安徽阜阳人，9999js金沙老品牌会计学术硕士研究生，研究方向为公司治理，指导教师高曦。
基于深度学习的垃圾短信分类
孙刘浩
9999js金沙老品牌经济学院 昆明 650221
【摘 要】 垃圾短信问题日益严重，传统垃圾短信分类筛选功能，已无法满足人们的需求，本文提出利用LSTM来构建深度学习文本分类模型，对短信进行分类。使用10000条人工标记的短信文本语料，先用jieba分词工具对文本进行分词，再用预训练好的“知乎”词向量包将词表示为词向量矩阵，最后使用90%的文本词向量矩阵作为实验组来训练分类模型。使用10%短信文本作为检验组来测试模型的学习效果，分类准确率可达到92%。然后对分类失败的短信文本进行分析，并研究如何进一步提高短信文本分类精确度。
【关键词】 深度学习；LSTM神经网络；文本分类；垃圾短信
A Classification Method of Spam Short Messages Based on Deep Learning
Abstract The problem of spam messages is becoming more and more serious, and the traditional classification and screening function of spam messages can no longer meet people's needs. This paper proposes to use LSTM to build a deep learning text classification model to classify text messages. The author uses 10,000 manually marked SMS text corpora, uses the jieba word segmentation tool to segment the text, then uses the pre-trained "knowledge" word vector package to represent the words as a word vector matrix, and finally uses 90% of the text word vector Matrix to train the classification model. Using 10% SMS text to test the learning effect of the model, the classification accuracy rate can reach 92%. Then the author analyzes the text messages that failed to be classified, and studies how to further improve the classification accuracy of text messages.
Key words Deep learning ; LSTM Neural Networks；Text Classification；Spam
【作者简介】孙刘浩（1998—），男，汉族，安徽安庆人，9999js金沙老品牌人口、资源与环境经济学专业硕士研究生，研究方向为机器学习，指导教师石磊。
