B1455
Title: Integrative multivariate regression analysis via penalization
Authors: Shuichi Kawano - Kyushu University (Japan) [presenting]
Toshikazu Fukushima - Nippon Steel Corporation (Japan)
Junichi Nakagawa - Nippon Steel Corporation (Japan)
Mamoru Oshiki - Hokkaido University (Japan)
Abstract: The multivariate regression models are widely used for analyzing data with multiple continuous responses and have been studied exhaustively. It offers the analysis of a single dataset. However, it is known that such a single-dataset analysis often leads to unsatisfactory results. Integrative analysis is an effective statistical approach to pool useful information from multiple independent datasets and provides better performance than single-dataset analysis. A multivariate regression model is proposed in integrative analysis. The integration is achieved by penalized estimation methods that perform variable and group selection. Numerical studies are conducted to examine the effectiveness of the proposed method.