讨论班 | ESL讨论班、学术交流研讨班、因果推断讨论班(2021/5/24-2021/5/30)

编辑:温夏玲 吴王威 许严方 责任审核人:谭键滨 蒋宇康

1、讨论班简介

ESL讨论班

针对新同学开展统计学习精要的学习,主讲机器学习知识。课程内容以《The Elements of Statistical Learning》为主,部分延伸内容需要参考辅助书目;

 

学术交流研讨班

针对博士生开展,主要形式为学术论文讨论交流。

 

因果推断讨论班

研究主题包括复杂数据的因果推断,精准治疗/决策,IV, DID, RDD, mediation等计量方法,异质性因果效应,因果机器学习,因果网络发现,非参检验,多源数据融合的因果分析和匹配、加权等经典方法的最新进展

 

2、时间及地点

 

image 16

 

 

3、本期内容概述

ESL讨论班

原型方法(Prototype)和最近邻方法是一种本质上不依赖具体模型,并且适用于分类和模式识别的方法。作为一种黑箱的预测器,它们在实际数据问题中的通常表现最好的模型之一。在本周的讨论班中,我们会重点介绍这两种方法。原型方法用特征空间中的点集来表示训练数据,当原型能够捕捉到每个类的分布时,方法将会比较有效,比如K均值聚类、量化学习向量和高斯混合模型等。而最近邻分类器不需要拟合模型,在目标点的k个最近邻样本点中采取少数服从多数的方法进行分类。进一步,我们介绍了自适应最近邻方法,根据目标点的环境情况调整决策区域,以获得误差更小的结果。

 

学术交流研讨班

由于网络通常是个体决策的结果,因此了解网络的形成对于研究网络效应非常重要。本文旨在使用网络结构上观察到的数据来识别和估计网络形成模型。我们将网络形成的特征描述为simultaneous-move game,其中形成边的utility取决于网络的结构,从而在边之间产生strategic interactions。由于多重均衡的普遍存在,我们使均衡的选择不受限制,并提出了一种通过借助观测到的子网络的partial identification方法。

 

因果推断讨论班

A review of spatial causal inference methods for environmental and epidemiological applications

Abstract:

The scientific rigor and computational methods of causal inference have had great impacts on many disciplines, but have only recently begun to take hold in spatial applications. Spatial casual inference poses analytic challenges due to complex correlation structures and interferencebetween the treatment at one location and the outcomes at others. In this paper, we review the current literature on spatial causal inference and identify areas of future work. We first discuss methods that exploit spatial structure to account for unmeasured confounding variables. We then discuss causal analysis in the presence of spatial interference including several common assumptions used to reduce the complexity of the interference patterns under consideration.These methods are extended to the spatiotemporal case where we compare and contrast the potential outcomes framework with Granger causality, and to geostatistical analyses involving spatial random fields of treatments and responses. The methods are introduced in the context of observational environmental and epidemiological studies, and are compared using both a simulation study and analysis of the effect of ambient air pollution on COVID-19 mortality rate. Code to implement many of the methods using the popular Bayesian software OpenBUGS is provided.