学术动态

罗季 | Causal Inference for Partial-Linear Model

时  间:2019年3月14日(周四)下午16:00-17:00

地  点:中北校区理科大楼A1716报告厅

题  目:Causal Inference for Partial-Linear Model

主讲人:罗季教授  浙江财经大学

摘  要:

Causal  inference is a subject developed based on statistics, which is devoted  to exploring causal relationship between things. It is widely used in  epidemiology, medicine, sociology, econometrics, behavioral science and  other disciplines. Since the reasoning method without modeling based on  the Bayesian conditional probability to the one with modeling, and then  to the one based on high-dimensional data, different methods of causal  reasoning depend on different data types. In the existing researches on  causal inference with modeling, most of them are used for analyzing  linear model and nonlinear model. Our work focused on the causal  inference for partially linear models, so as to extend largely the  application of causal inference in more fields. We proposed an estimate  method of the unknown parameters and the undiscovered functions.  Furthermore, we studied the influence of confounding variable on the  robustness of causal inference in partial linear models. In the end, we  applied our theoretical results to causal analysis for  solitary-papillary thyroid carcinoma metastasis data.

报告人简介:

罗季,浙江财经大学数据科学学院副院长,教授,博士生导师。浙江省高校中青年学科带头人,中国现场统计研究会大数据统计分会常务理事、生存分析分会理事,浙江省统计学会理事。主要从事大数据分析、半参数统计推断、因果推断、统计计算等领域的研究。近年来,先后访问美国南卡罗莱纳大学生物统计系,加拿大不列颠哥伦比亚大学统计系。主持国家级、省部级项目6项,发表论文30余篇。

发布者:张瑛发布时间:2019-03-11浏览次数:81