统计与决策2025，Vol.41Issue(22)：11-16,6.DOI:10.13546/j.cnki.tjyjc.2025.22.002
基于树状模型的分位数处理效应估计研究
Research on Quantile Treatment Effect Estimation Based on Tree Model
摘要
Abstract
In view of the fact that causal tree and causal forest can only be used to estimate the average treatment effect and cannot obtain the heterogeneity of treatment effect,and that by taking advantage of the characteristic that a well-performing tree model can capture the complex relationships among variables and improve the accuracy of predictions,this paper starts from the definition of quantile treatment effect,and uses regression tree and random forest to predict the potential outcome of individuals,to estimate the quantile treatment effect,and to describe the heterogeneity of the outcome variable at different quantiles,which im-proves the deficiency of causal tree and causal forest in the treatment effect analysis.关键词
分位数处理效应/回归树/随机森林Key words
quantile treatment effect/regression tree/random forest分类
社会科学引用本文复制引用
赵文丽,石洪波..基于树状模型的分位数处理效应估计研究[J].统计与决策,2025,41(22):11-16,6.基金项目
教育部人文社会科学研究规划基金项目(22YJAZH092) （22YJAZH092）
