香蕉视频

Enriched post-selection models for high dimensional data

High dimensional data are rapidly growing in many domains, for example, in microarray gene expression studies, fMRI data analysis, large-scale healthcare analytics, text/image analysis, natural language processing and astronomy, to name but a few. In the last two decades regularisation approaches have become the methods of choice for analysing high dimensional data. However, obtaining accurate estimates and predictions as well as valid statistical inference remains a major challenge in high dimensional situations. In this talk, we present enriched post-selection models that aim to improve parameter estimation and prediction, and to facilitate statistical inferences in high dimensional regression models. The enriched post-selection method enables us to construct valid post-selection inference for regression parameters in high dimensions. We discuss the empirical and asymptotic properties of the enriched post-selection method.

Dr. Reza Drikvandi is an Assistant Professor of Statistics from the Department of Mathematical Sciences at Durham University.

His research mainly focuses on high dimensional statistics, longitudinal data analysis, mixed-effects models, nonparametric and semiparametric models, joint modelling, model diagnostics and missing data problems.

McGill Statistics Seminar schedule:

Meeting ID: 834 3668 6293

Passcode: 12345

听