|
Activity Number:
|
441
- Novel Statistical and Machine Learning Approaches for Business and Financial Services
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 6, 2020 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
Section on Statistical Computing
|
|
Abstract #312749
|
|
|
Title:
|
Recent Developments on Multivariate Graph Constrained Models
|
|
Author(s):
|
Brad Price* and Ben Sherwood
|
|
Companies:
|
West Virginia University and University of Kansas
|
|
Keywords:
|
Multivariate;
Penalized Likelihood;
Inference
|
|
Abstract:
|
In this talk we will introduce a new framework for sparse multivariate regression which uses a minimum penalty to relate responses through the mean function. We provide computational and theoretical results for this new model, as well and a comparison to standard competitors. As this is a specific case of a multivariate graph constrained model, we also provide insights on post selection inference for more general settings. Finally we will present an example admission to addiction facilities related to the opioid crisis. This is joint work with Ben Sherwood from the University of Kansas.
|
Authors who are presenting talks have a * after their name.
