Activity Number:

453
 Novel Theory and Methods in Big Data Analytics

Type:

Invited

Date/Time:

Wednesday, August 1, 2018 : 8:30 AM to 10:20 AM

Sponsor:

Section on Statistical Learning and Data Science

Abstract #326781

Presentation

Title:

Complex Interaction Modeling with Liquid Association

Author(s):

KerChau Li*

Companies:

Institute of Statistical Science, Academia Sinica

Keywords:

sliced inverse regression;
liquid association

Abstract:

Multivariate regression aims at the study of the relationship between one set of input variables X and one set of output variables Y. Challenges occur when no parametric models are known and yet the number of variables is large. To overcome the difficulties, dimension reduction methods under the inverse regression viewpoint have been investigated by many authors. Liquid association (LA) depicts the change in the covariation of two variables X and Y as a third variable Z varies. It was introduced in Li(2002) in the context of gene expression data analysis. In this talk, I will describe a framework to illustrate how the LA methodology can help increase the modeling flexibility of multivariate regression in analyzing complex data.
