因果链上因果效应的关系及推断

探讨因果链上总因果效应与局部因果效应的关系及因果效应的可识别性. 给出了在因果链上因果效应的传递关系. 根据这个关系, 给出了通过控制因果链上中间因素消除混杂偏倚的方法.     

一个虚拟事实模型中因果效应的Bayes估计

给出了一个虚拟事实模型中因果效应的Bayes估计和经验Bayes估计,提供了三种可替换性假设的先验分布的选择方法,并用实验说明,在不知道取哪个可替换性假设的情况下,经验Bayes估计要优于其他的估计.     

基于扩展事件图的因果分析方法及在仿真中的应用

通过因果分析对仿真结果作出因果解释是仿真非常重要的一个环节.为了能够支持因果分析,首先提出扩展事件图,以重构仿真中发生的事件以及事件之间的因果关系;然后根据扩展事件图提出若干分析算法,通过对图模型的概率参数的可能性和主要性分析,获取因果知识;最后以一个攻防对抗仿真为例,对以上方法进行了必要的验证.     

Granger因果检验的非线性进展及应用研究

Granger因果检验是计量经济学的重要组成部分,也是现代经济、金融学分析的重要工具,近年来Granger因果检验在非线性检验方向有了较大进展。本文在线性Granger因果检验的基础上,阐述了Granger因果检验的非线性进展,重点总结了针对一阶矩的基于回归模型、非参函数和信息理论的三大类非线性方法以及针对二阶矩的基于残差交叉相关系数和多元条件方差模型下的两大类非线性方法,讨论了不同非线性Granger方法中数据要求、核心模型、建模关键以及模型优缺点,提出了Granger因果检验"线性-非线性"的整体框架和研究范式.通过模型分析和比较,本文可为因果检验的非线性理论和模型研究提供参考,并对因果检验在经济和金融领域的更广泛应用提供支持。     

基于多变量的Granger因果检验方法

针对既有的Granger因果检验方法只能处理两个变量间的因果关系问题,指出这种方法会导致间接因果与直接因果的混淆及由于数据同源而产生的伪因果问题,提出一种可以消除以上问题的多变量因果检验方法,该方法立足于原有的Granger因果检验,适用于短时相依的和平稳的时间序列数据,并根据蒙特卡罗方法给出了统计推断的检验量,设计了方法的实施步骤.最后,应用一个仿真的实例具体展示了方法的使用过程和方法的有效性。     

一类因果模型的可识别性条件

因果问题在近代医学,生物学,社会科学的研究中占有非常重要的地位。通过因果关系预见某些行为或策略对研究对象的影响已经成为一些实际研究的最终目的。Rubin(1978)提出了解决因果问题的虚拟事实模型,建立了因果推断统计分析的基本框架。虚拟事实模型的因果效应是以实际观测数据为研究对象的,但又不完全由数据之间的相关性决定,因此在讨论因果效应时存在可识别性问题。如果因果效应可识别,则有可能利用观测数据直接计算因果效应。但是,众 所周知:在不加任何假设或限制的条件下,虚拟事实模型的因果效应是不可识别的。若要研究变量间的因果效应就必须对虚拟事实模型加入某些必要的限制,使因果效应在这些限制下可识别。郑忠国,张艳艳,童行伟在“因果模型因果效应的可识别性研究”中针对控制变量与协变量相互独立的一类模型的可识别性进行了研究,指出在某些特定的可替换性假设之下,模型的因果效应具有可识别性。该文将针对控制变量作用于协变量的虚拟事实模型进行可识别性研究。作者将指出:控制变量是否作用于协变量并不影响因果效应的可识别性和可替换性假设。并给出:此类模型因果效应可唯一确定的充要条件 。      

带有反馈的因果模型中的独立性识别

在本文中,直接利用计算要概率分布的办法证明了在史包含离散变量４ 反馈系统产生的因果图中的条件独立关系可以由ｄ－分离识别出．     

非平稳过程的小波因果度量

该文对带有随机趋势的非平稳过程给出了一种新的因果度量定义.作者将证明新定义与Hosoya给出的因果度量定义等价, 但由于避免了以往文献中常见的对非平稳过程协整性的要求, 该定义有利于简化因果关系的假设检验.文中还对Wald检验和似然比检验进行了讨论.数值模拟和实证分析表明, 这两种检验方法都是有效的.     

基于NEGM(1,1)模型的因果预测方法及应用研究

非等间隔时间序列在工程技术问题中是常见的.研究了一类非等间隔广义时间序列的预测问题,也就是将因果预测模型中的自变量作为广义时间,应用NEGM(1,1)模型将因果预测转化为时间序列预测,并应用空军航材消耗实例进行了模型检验.实践表明本文的方法具有广泛的使用价值.     

因果模型中因果效应的可识别性研究

对有向非循环图中的因果关系进行探讨,提出了相应的虚拟事实模型,并利用条件独立关系作为辅助信息对一种较为简单的虚拟事实模型的可识别性进行了研究.将可忽略假设推广为可替换性假设,并且得到：在可替换性假设之下,因果效应具有可识别性,即有可能从观察数据直接计算因果效应.     

Causal statistical inference in high dimensions

We present a short selective review of causal inference from observational data, with a particular emphasis on the high-dimensional scenario where the number of measured variables may be much larger than sample size. Despite major identifiability problems, making causal inference from observational data very ill-posed, we outline a methodology providing useful bounds for causal effects. Furthermore, we discuss open problems in optimization, non-linear estimation and for assigning statistical measures of uncertainty, and we illustrate the benefits and limitations of high-dimensional causal inference for biological applications.     

Identifying the Causal Effects of Stochastic Interventions by Augmented Causal Networks

Suppose that the cause-effect relationships between variables can be described by a causal network. To identify the causal effect of a stochastic intervention, an augmented causal network for stochastic intervention is proposed in this paper. Then, we obtain two graphical criteria for identifying the causal effects of stochastic interventions from passive observations on observed variables only. When either of the two criteria is satisfied, a simple closed-form expression is provided for the causal effect of a stochastic intervention, which enables researchers to assess the causal effect with little effort.     

Identifiability of intermediate variables on causal paths

We discuss the discovery of causal mechanisms and identifiability of intermediate variables on a causal path. Different from variable selection, we try to distinguish intermediate variables on the causal path from other variables. It is also different from ordinary model selection approaches which do not concern the causal relationships and do not contain unobserved variables. We propose an approach for selecting a causal mechanism depicted by a directed acyclic graph (DAG) with an unobserved variable. We consider several causal networks, and discuss their identifiability by observed data. We show that causal mechanisms of linear structural equation models are not identifiable. Furthermore, we present that causal mechanisms of nonlinear models are identifiable, and we demonstrate the identifiability of causal mechanisms of quadratic equation models. Sensitivity analysis is conducted for the identifiability.     

Relationship of causal effects in a causal chain and related inference

This paper discusses the relationship among the total causal effect and local causal effects in a causal chain and identifiability of causal effects. We show a transmission relationship of causal effects in a causal chain. According to the relationship, we give an approach to eliminating confounding bias through controlling for intermediate variables in a causal chain.     

Identifiability of Dynamic Systems Represented by Bond Graphs

In this paper, the identifiability of dynamical systems is presented from the perspective of the Bond Graph Methodology. Several classical approaches for identifiability are studied and their applicability to bond graphs is discussed. The Markov Parameter method was selected to determine local identifiability of linear systems and a methodical procedure to obtain Markov Parameters directly from the bond graph is presented which is based on its causal paths. This method can easily be implemented in on a computer and works for bond graphs with derivative causality elements and fields.     

Causal emergentism

In this paper I describe basic features of traditional (British) emergentism and Popper’s emergentist theory of consciousness and compare them to the contemporary versions of emergentism present in connectionist approach in cognitive sciences. I argue that despite their similarities, the traditional form, as well as Popper’s theory belong to strong causal emergentism and yield radically different ontological consequences compared to the weaker, contemporary version present in cognitive science. Strong causal emergentism denies the causal closure of the physical domain and introduces genuine new mental causal powers and genuine downward causation, while weak emergentism provides new insights in understanding the mechanisms and explanation that is compatible with physicalism.     

可压缩磁流体动力方程解的正则性

该文研究三维等熵磁流体动力方程和二维带正密度的热传导磁流体动力方程解的正则性.给出了局部强解爆破的条件.     

Causal Truthmaking

This paper provides an outline of a theory of causal truthmaking according to which contingent truths are made true by causal facts and dispositional mechanisms. These facts and mechanisms serve to account for the truth of propositions by explaining in a non-epistemic fashion why they have come about as truths. Given that negative causation is allowed for, we are able to provide truthmakers for negative truths without making appeal to negative facts, lacks or absences. The paper takes its starting point in the following claims by George Molnar: (1) the world is everything that exists; (2) everything that exists is positive; (3) some negative claims about the world are true; and (4) every true claim about the world is made true by something that exists. The conclusion is reached that we can keep (1–4) in a consistent manner if causal truthmaking is permitted.     

Identifiability of causal effect for a simple causal model

Counterfactual model is put forward to discuss the causal inference in the directed acyclic graph and its corresponding identifiability is thus studied with the ancillary information based on conditional independence. It is shown that the assumption of ignorability can be expanded to the assumption of replaceability,under which the causal efiects are identifiable.