SPHIS Home » Departments » Bioinformatics & Biostatistics » Research » Seminar Series » 2024-09-13
2024-09-13
Peng Wang, Ph.D., College of Business, University of Cincinnati
"Repro Sampling Method and Its Application on Inference of High Dimensional Linear Models"
In this talk, we first present a novel, general and effective simulation-based approach, called Repro Sampling method, to conduct statistical inference by creating and studying the performance of artificial samples generated by mimicking the true observed sample. The artificial samples, referred to as repro samples, are used to quantify uncertainty and provide confidence sets with guaranteed coverage rates for the target parameter or quantity on a wide range of problems, including many where solutions were previously unavailable or could not be easily obtained. We then present a general theoretical framework and an effective Monte-Carlo algorithm, with supporting theories, for inference of high dimensional linear models. This method is used to create confidence sets for both the selected models and model coefficients, with both exact and asymptotic inferences, are included. It also provides theoretical development to support computational efficiency. The development provides a simple and effective solution for the difficult post-selection inference problems.
