The G-estimate of the regularized Mahalanobis distance in the case where the distribution of observations is different from the normal one. (Russian. English summary)

• G-estimation of the regular Mahalanobis distance under the condition that the dimension of the sample space is comparable with the size of the learning samples is analysed. It is proved for the first time that properties of consistency and asymptotic normality of this estimation hold, if the distribution of observations is not normal.

Subject headings

NATURVETENSKAP  -- Matematik (hsv//swe)

NATURAL SCIENCES  -- Mathematics (hsv//eng)

Keyword

non-normal observations

G-estimation of the regular Mahalanobis distance

learning samples

consistency

asymptotic normality

MATHEMATICS

MATEMATIK

Publication and Content Type

ref (subject category)

art (subject category)