RUS  ENG
Full version
JOURNALS // Avtomatika i Telemekhanika // Archive

Avtomat. i Telemekh., 1986 Issue 7, Pages 61–69 (Mi at6380)

Stochastic Systems

Optimality of classification of random observations differing in their regression equations

V. I. Malyugina, Yu. S. Kharin

a Minsk

Abstract: The problem of classifying multidimensional observations different in the regression equations is considered for the case when the parameters of the regression equations are not available in advance and are estimated from a random sample (classified or not). Probability error estimates are provided for decision rules with learning and selflearning, depending on the sample size and the experiment design. A numerical example is given.

UDC: 62-505, 519.272


Received: 24.06.1985


 English version:
Automation and Remote Control, 1986, 47:7, 926–935

Bibliographic databases:


© Steklov Math. Inst. of RAS, 2024