mailto:uumlib@uum.edu.my 24x7 Service; AnyTime; AnyWhere

Strategies for non-parametric smoothing of the location model in mixed-variable discriminant analysis

Mahat, Nor Idayu and Krzanowski, W.J. and Hernandez, A. (2009) Strategies for non-parametric smoothing of the location model in mixed-variable discriminant analysis. Modern Applied Science, 3 (1). pp. 151-163. ISSN 1913-1852

[thumbnail of Str.pdf] PDF
Restricted to Registered users only
Available under License Creative Commons Attribution.

Download (216kB)

Abstract

The non-parametric smoothing of the location model proposed by Asparoukhov and Krzanowski (2000) for allocating objects with mixtures of variables into two groups is studied. The strategy for selecting the smoothing parameter through the maximisation of the pseudo-likelihood function is reviewed. Problems with previous methods are highlighted, and two alternative strategies are proposed. Some investigations into other possible smoothing procedures for estimating cell probabilities are discussed. A leave-one-out method is proposed for constructing the allocation rule and evaluating its performance by estimating the true error rate. Results of a numerical study on simulated data highlight the feasibility of the proposed allocation rule as well as its advantages over previous methods, and an example using real data is presented.

Item Type: Article
Uncontrolled Keywords: Brier score, Error rate, Leave-one-out process, Location model, Pseudo-likelihood
Subjects: Q Science > QA Mathematics > QA76 Computer software
Divisions: College of Arts and Sciences
Depositing User: Mrs. Norazmilah Yaakub
Date Deposited: 25 Feb 2012 03:06
Last Modified: 25 Feb 2012 03:06
URI: https://repo.uum.edu.my/id/eprint/4610

Actions (login required)

View Item View Item