Lu Tang, of the Department of Biostatistics at the University of Pittsburgh, will present:
“Leveraging Models from Heterogeneous Data Sources to Improve Personalized Treatment Effect Estimation”
Abstract: Individualized causal inference, ranging from personalized medicine to marketing advertisement, has remained a hot topic. However, due to the limited sample size in a single study, estimating treatment effects is often challenging. Here we propose a tree-based model averaging approach to improve the estimation efficiency of conditional average treatment effects (CATE) concerning the population of a targeted research site by leveraging models derived from potentially heterogeneous populations of other sites, but without them sharing individual-level data. Drawing on a multi-hospital electronic health records network, we develop an efficient and interpretable tree-based ensemble of personalized treatment effect estimator to join results across hospital sites, while actively modeling for the heterogeneity in data sources through site partitioning. The performance of this approach is demonstrated by a study of causal effects of oxygen saturation on hospital mortality and backed up by comprehensive numerical results. Causal assumptions with practical implications are also explored to warrant the use of our approach. This is joint work with Xiaoqing Tan and Joyce Chang at Pitt.
All are Welcome.