JSM 2015 Preliminary Program

Online Program Home
My Program

Abstract Details

Activity Number: 655
Type: Contributed
Date/Time: Thursday, August 13, 2015 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Learning and Data Mining
Abstract #315969
Title: A Regularized Approach to Sparse Linear Discrimination Analysis for Two-Class Classification
Author(s): Angang Zhang* and Xinwei Deng
Companies: and Virginia Tech
Keywords: Two-class classification ; sparse linear discriminate analysis ; misclassification error ; extended BIC
Abstract:

Linear discriminate analysis (LDA) is a commonly used method for two-class classification. However, LDA may not perform well in the case of high dimensional data with number of variables p larger than the number of observations n. In this work, we proposed a novel sparse LDA which generalizes LDA through a regularized approach with shrinkage on both the inverse covariance matrix and the mean difference between the two classes. The proposed method enjoys the advantage of ease of interpretation with efficient computation. Simulation under multiple data generation settings are conducted to compare the proposed method with other commonly used methods in terms of estimation accuracy and misclassification error. The performance of proposed method is also examined through real leukemia and lung cancer data.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2015 program





For program information, contact the JSM Registration Department or phone (888) 231-3473.

For Professional Development information, contact the Education Department.

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

2015 JSM Online Program Home