On Group Choice Procedures for Problems of Classification and Reliability in the Case of Lognormal Variance*

Autor:	R. A. Abusev
Rok vydání:	2013
Předmět:	Statistics and Probability Applied Mathematics General Mathematics Bayesian probability Estimator Variance (accounting) Decision rule Linear inequality Group analysis Robustness (computer science) Statistics Log-normal distribution Econometrics Mathematics
Zdroj:	Journal of Mathematical Sciences. 189:911-918
ISSN:	1573-8795 1072-3374
DOI:	10.1007/s10958-013-1231-y
Popis:	This paper is the first to obtain results by applying group analysis to the problems of statistical estimation, statistical pattern recognition, and robustness in the case of populations with lognormal distributions. Maximum likelihood unbiased Bayesian estimators are constructed, a variety of decision rules for group recognition is developed, and their statistical properties and qualitative characteristics are examined. Analytical expressions are found for the probability of the linear stochastic inequality as well as various statistical estimates for this probability. The relations between recognition problems and the probability of a linear inequality are analyzed.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::919e8fbb006db9b5fec3795908df6926 https://doi.org/10.1007/s10958-013-1231-y Zobrazit plný text záznamu Plný text ve formátu PDF