A Procedure for Estimating the Number of Clusters in Logistic Regression Clustering |
| |
Authors: | Guoqi Qian Yuehua Wu Qing Shao |
| |
Institution: | (1) Department of Mathematics and Statistics, The University of Melbourne, Melbourne, VIC, 3010, Australia;(2) Department of Mathematics and Statistics, York University, Toronto, ON, M3J 1P3, Canada;(3) Biostatistics and Statistical Reporting, One Health Plaza, Bldg. 435–4173, Novartis Pharmaceuticals Corporation, East Hanover, NJ 07936, USA |
| |
Abstract: | This paper studies the problem of estimating the number of clusters in the context of logistic regression clustering. The
classification likelihood approach is employed to tackle this problem. A model-selection based criterion for selecting the
number of logistic curves is proposed and its asymptotic property is also considered. The small sample performance of the
proposed criterion is studied by Monto Carlo simulation. In addition, a real data example is presented.
The authors would like to thank the editor, Prof. Willem J. Heiser, and the anonymous referees for the valuable comments and
suggestions, which have led to the improvement of this paper. |
| |
Keywords: | Asymptotics Logistic regression clustering Model selection Penalty |
本文献已被 SpringerLink 等数据库收录! |
|