Seminar: Envelope Models and Methods

2015-06-24  Xiaodong Pan Hits:[]

Topic: Envelope Models and Methods

Speaker: Dr. Zhihua Su

Abstract:

This talk presents an introduction of a recent statistical concept called an envelope. An envelope has the potential to achieve substantial efficiency gains in multivariate analysis by identifying and accounting for immaterial information in the data. The efficiency gains are demonstrated both by theory and examples. Today the scope of envelopes is extended well beyond multivariate linear regression.  An overview of recent developments in this area will be discussed. Applications of envelopes and their connection to other fields will also be mentioned.

Time:9:00--10:00, June 29, 2015

Room:X2511

 

Profile:Dr. Zhihua Su is an assistant professor at Department of Statistics, University of Florida (USA). She got her Ph.D. in statistics for University of Florida in 2012. Her research interests mainly lies in Dimension reduction, Multivariate analysis, Bioinformatics, Model Selection, Nonparametric statistics.

Pre:Seminars: Sparse Estimation via Approximated Information Criteria Next:Seminar: Testing Multiple Hypothesis in Big Data Analysis

Close

Baidu
sogou