学术动态

Doubly Robust Survival Trees and Random Forests

报 告 人: 刁力群

报告地点: 数学与统计学院四楼报告厅

报告时间: 2016年12月14日星期三15:30-16:30

报告简介:

Estimating a patient’s mortality risk is important in making treatment decisions. Tree-based methods are useful tools to identify risk groups and conduct prediction by employing recursive partitioning to separate patients into different risk groups. Existing “loss based" recursive partitioning procedures that would be used in the absence of censoring have previously been extended to the setting of right censored outcomes using inverse probability censoring weighted estimators of loss functions. We propose new "doubly robust" extensions of these loss estimators motivated by semi-parametric efficiency theory for missing data that better utilize available data. We realized such extensions by imputation and extended this single tree method to doubly robust survival random forests (ensemble methods).

 

主讲人简介:

Liqun obtained her Ph.D. degree at Univ. of Waterloo in 2013 and her Ph.D. thesis won 2013 Pierre Robillard Award from the Statistical Society of Canada (The best doctoral thesis in probability or statistics defended at a Canadian university in 2013). She then went to Univ. of Rochester to do a postdoc. She started to work as an assistant professor at Univ. of Waterloo since July 2015.

 

专题网站Project site