After successful completion of the course, students are able to have a solid understanding of Sufficient Dimension Reduction and to use the SDR methods developed over the last three decades in order to analyze multivariate data. Furthermore they are able to summarize a specific SDR method and give a presentation about it.
Sufficient Dimension Reduction (SDR) aims to find sufficient, in the statistical sense, reductions of a large set of explanatory variables in order to model a target response Y. The reduction and targeting are carried out simultaneously as SDR identifies a sufficient function of the regressors X that preserves the information in the conditional distribution of Y given X.
The course will cover most moment and model based SDR methods and recent methodological developments. The students will study instructor provided material and will present papers during the course.
This is a seminar course. The instructor will present an introduction and overview of the topic and students will present papers.
Students will present papers on topics in sufficient dimension reduction.