美中贸易协调性的实证分析:垂直专业化视角

基于2005～2012年的月度数据进行实证分析,协整检验结果表明,美国对华出口与从中国进口存在长期稳定的协整关系。从误差修正模型看,短期内美国对华出口与从中国进口之间存在动态调整机制。根据格兰杰因果关系检验,在滞后6期内,美国对华出口与从中国进口之间只存在单向的因果关系,美国对华出口是美国从中国进口的格兰杰原因,美国对华出口增长是美国从中国进口增长的决定因素。脉冲响应函数分析表明,美国对华出口的某一冲击会给美国从中国进口增长带来同向冲击。这些实证结果表明垂直专业化分工在驱动着美中贸易,政府在制定平衡美中贸易的政策时,不应忽视垂直专业化分工以及市场对维持美中双边贸易均衡的约束机制。     

刑法因果关系之判断——兼论司法实践中的操作

我国刑法因果关系理论应当摆脱必然与偶然之争,借鉴国外因果关系的判断理论丰富我国的研究并为司法实践服务。以条件关系为基础判断事实因果关系,以相当因果关系说为基础评价法律因果关系,以客观归属论为借鉴为相当性提供明确的标准。     

Ambient Intelligence, Criminal Liability and Democracy

In this contribution we will explore some of the implications of the vision of Ambient Intelligence (AmI) for law and legal philosophy. AmI creates an environment that monitors and anticipates human behaviour with the aim of customised adaptation of the environment to a person’s inferred preferences. Such an environment depends on distributed human and non-human intelligence that raises a host of unsettling questions around causality, subjectivity, agency and (criminal) liability. After discussing the vision of AmI we will present relevant research in the field of philosophy of technology, inspired by the post-phenomenological position taken by Don Ihde and the constructivist realism of Bruno Latour. We will posit the need to conceptualise technological normativity in comparison with legal normativity, claiming that this is necessary to develop democratic accountability for the implications of emerging technologies like AmI. Lastly we will investigate to what extent technological devices and infrastructures can and should be used to achieve compliance with the criminal law, and we will discuss some of the implications of non-human distributed intelligence for criminal liability.

立法伦理与算法正义——算法主体行为的法律规制

人类设计算法、运用算法的最终目的是为了增强自身解决问题的能力,进一步提升生活水平并优化社会结构。算法黑箱存在的前提预设是人类对因果关系的认知断裂,算法黑箱针对自动化决策过程中特定的算法用户。在一系列的数据代码背后是算法主体的行为目标和价值选择,不同价值选择加上算法的涌现性和自主性的复杂特质导致相关法律关系主体的权利冲突。人类的有限理性是算法底层逻辑的道德基础,算法的底层指令和代码归根结底是人类的思考方式的展现,在这一逻辑前提下,为算法立法的过程就还原为对于算法主体行为的法律规制问题。立法伦理为法律关系主体行为的合法性和合理性提供了德性基石,算法主体的理性认知要符合立法伦理要求。从立法原则、立法技术、程序机制和价值规范等方面规制算法主体的行为,最终迈向算法正义。     

论“网络暴力”致人自杀死亡的刑事责任

随着互联网科技的发展,虚拟空间与现实生活的联系越来越紧密,网络暴力事件层出不穷,出现了不少因不堪忍受凌辱而自杀的案例。我国刑法通常认为自杀死亡结果和网络暴力之间不存在因果关系,自杀死亡结果只是情节严重中的结果,但这并不合理。网络暴力的特质在于对精神的强制以及对身心的持续伤害,网络暴力高于普通的精神伤害行为,在特定的情况下网络暴力能成为杀人罪的实行行为并与自杀死亡结果有因果关系。相应地,刑事责任也应重新评价。     

山东省技术进步与地区经济增长之间关系的实证分析

在地区经济增长中,技术进步是一个重要因素,自改革开放来山东省技术进步与经济增长的关系有待考察。首先采用格兰杰因果关系检验方法,得出二者存在双向因果关系;然后通过ADF检验,利用协整检验方法,说明山东省的经济增长与技术进步不存在长期稳定的关系。     

Macro-determinants of the Lithuanian housing market: a test for Granger causality

This research examines the macro-determinants of the Lithuanian housing market. The study employs the Granger causality test to assess the interdependence of a number of macro-variables and the national housing price index. Regardless of the limitations involved with this methodology, the empirical findings suggest that the Lithuanian housing market relates to growth in building activity, interest rates, inflation, and employment. Considering all of these results, the research highlights useful policy implications for property market participants. Investors and developers can employ this information to guide their investment decisions. Likewise, the government and central bank could use this updated knowledge to drive a successful macroeconomic program.     

区域金融生态环境与金融资源配置的经验分析——以安徽省为例

运用典型相关分析与Granger因果检验等方法对安徽省区域金融生态环境与金融资源配置水平之间进行实证分析。两者存在Granger因果关系。但这种关系是单向的,是从金融生态环境到金融资源配置,而且表现为正向。建议从政府主导型市场经济向政府监管型市场经济的过渡,使两者相互促进的作用得以充分发挥。     

Impact evaluations in South Korea and China

While evidence-based policy-making is increasingly in demand, as new policies are required to bring effective results to targeted groups in South Korea and China, few studies have investigated the progress of quantitative impact evaluation that focuses on causality. This paper studies the trends of quantitative impact evaluation of public policy in South Korea and China by surveying major public administration and public policy journals in these two countries from 2000 to 2015. Among published articles in the major journals, our study pool includes research articles directly related to quantitative impact evaluation. Our study found that there has been considerable progress in impact evaluation research in South Korea and China in both data quality and empirical methods. However, empirical impact evaluation still comprises a small fraction (only one to two percent) of all research in public administration and public policy in both countries. We also found limited discussion on the selection mechanism and related bias in South Korea even in recent years, while causality and selection bias have been more commonly discussed in China. Also, advanced empirical methods are more frequently observed in journal articles in China than those in South Korea.