Comparison of two estimators of mean function in LDA of spatially correlated Gaussian data

Abstract

The problem of linear discriminant analysis of an observation of Gaussian random field into one of two populations is considered. In this paper we analyze the performance of the plug‐in linear discriminant function, when unknown means are estimated from the training samples. The generalized least squares and the ordinary least squares estimators are used. Obtained asymptotic expansions for the expected error rate are compared numerically in the case of spherical models for population covariances.

Straipsnyje sprendžiamas atsitiktinio Gauso lauko stebejimu tiesines diskriminantines analizes uždavinys dvieju klasiu atveju. Gauti pirmos eile asimptotiniai tiketinos klasifikavimo klaidos skleidiniai atvejui, kai i Bajeso klasifikavimo taisykle istatome maksimalaus tiketinumo bei empirini vidurkiu iverčius. Atliktas skaitinis asimptotiniu klasifikavimo klaidu palyginimas tam tikrai kaimynystes schemai bei sferinei koreliaciju funkcijai.