Atlas home || Conferences | Abstracts | about Atlas

International Conference on Statistics, Combinatorics and Related Areas - 7th International Conference of the Forum for Interdisciplinary Mathematics
December 19-21, 2000
Indian Institute of Technology-Bombay
Mumbai, Maharastra, India

Organizers
Satya N. Mishra (University of South Alabama), Sanjeev V. Sabnis (IIT, Bombay)

View Abstracts
Conference Homepage

SIRP: An extended SIR algorithm based on principal points
by
Megu Ohtaki
Research Institute for Radiation Biology and Medicine, Hiroshima University, JAPAN
Coauthors: Li-Xing Zhu (University of Hong Kong, CHINA)

Sliced inverse regression (Li, 1991) is known as a powerful tool for estimating them. However, it is well known that when dependence is symmetric, the SIR is not efficient (Li 1991). In this article, we propose a modified SIR algorithm to fill the leak of the original SIR algorithm. The proposed algorithm, which is based on pricipal point notion, have good performance in either case of symmetric or asymmetric dependence. Simulation study is performed for illustration.

Date received: November 6, 2000

Copyright © 2000 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cafr-94.