ON NONPARAMETRIC MULTIVARIATE BINARY DISCRIMINATION
成果类型：
Article
署名作者：
HALL, P
刊物名称：
BIOMETRIKA
ISSN/ISSBN：
0006-3444
DOI：
10.2307/2335829
发表日期：
1981
页码：
287294
关键词：
摘要：
Aitchison and Aitken (1976) introduced a novel nonparametric method for estimating probabilities in a multidimensional binary space. The technique is designed for use in multivariate binary discrimination. Their estimator [applied to the data of Anderson et al. (1972) on the diagnosis of keratoconjunctivitis sicca by 10 different symptoms] depends crucially on an unknown smoothing parameter .lambda.; Aitchison and Aitken proposed a maximum likelihood method for determining .lambda. from the sample. This leads to an adaptive estimator which can behave erratically when there are empty or near empty cells present. This was both theoretically and by example. To overcome these difficulties another method was introduced of estimating .lambda., which is designed to minimize a global function of the mean squared error.