EE Systems Seminar
Estimation after Parameter Selection
Abstract In many practical parameter estimation problems, such as medical experiments and cognitive radio communications, parameter selection is performed prior to estimation. The selection process has a major impact on subsequent estimation by introducing a selection bias and creating coupling between decoupled parameters. As a result, classical estimation theory may be inappropriate and inaccurate and a new methodology is needed. In this study, the problem of estimating a preselected unknown deterministic parameter, chosen from a parameter set based on a predetermined data-based selection rule, Ψ, is considered. In this talk, I will present a general non-Bayesian estimation theory for estimation after parameter selection, includes estimation methods, performance analysis, and adaptive sampling strategies. The new theory is based on the post-selection mean-square-error (PSMSE) criterion as a performance measure instead of the commonly used mean-square-error (MSE). We derive the corresponding Cramér-Rao-type bound on the PSMSE of any Ψ-unbiased estimator, where the Ψ -unbiasedness is in the Lehmann-unbiasedness sense. Then, the post-selection maximum-likelihood (PSML) estimator is presented and its Ψ–efficiency properties are demonstrated. Practical implementations of the PSML estimator are proposed as well. As time permits, I will discuss the similar ideas that can be applied to estimation after model selection and to estimation in Good-Turing models.
Bio Tirza Routtenberg (S'07-M'13-SM'18) received the B.Sc. degree (magna cum laude) in bio-medical engineering from the Technion Israel Institute of Technology, Haifa, Israel in 2005 and the M.Sc. (magna cum laude) and Ph.D. degrees in electrical engineering from the Ben-Gurion University of the Negev, Beer-Sheva, Israel, in 2007 and 2012, respectively. She was a postdoctoral fellow with the School of Electrical and Computer Engineering, Cornell University, in 2012-2014. Since October 2014, she is a faculty member at the Department of Electrical and Computer Engineering, and Ben-Gurion University of the Negev, Beer-Sheva, Israel. Her research interests include signal processing in smart grid, statistical signal processing, estimation and detection theory, and signal processing on graphs. She was a recipient of the Best Student Paper Award in International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2011, in IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP) 2013 (coauthor), in ICASSP 2017 (coauthor), and in IEEE Workshop on Statistical Signal Processing (SSP) 2018 (coauthor).
Contact: Liliana Chavarria at 626-395-4715 email@example.com