Statistical inference for high-dimensional regression with proxy data

Speaker: Associate professor LI Sai, Renmin University of China

Time: 14:00, March 8

Venue: Room 1417, Administrative Building, Zijingang Campus

Abstract: We study estimation and inference for high-dimensional linear models with two types of “proxy data”. The first type of proxies encompasses marginal statistics and sample covariance matrices computed from distinct sets of individuals. We develop a rate optimal method for estimation and inference for the regression coefficient vector and its linear functionals based on the proxy data. We show the intrinsic limitations in the proxy-data based inference: the minimax optimal rate for estimation is slower than that in the conventional case where individual data are observed. The second type of proxy data is differentially private data. We propose method for private estimation and inference in high-dimensional regression with FDR control.