Multi-Classification Logistics Regression In Ranked Set Sampling And Its Application

In this paper, we study the parameter estimation problem of ranked set sampling(RSS) under the multi-classification logistics regression model. In the existing research literature, we use the method of ranked set sampling of several steps binary logistic regression and then use the data to estimate the parameters. However, in the case of a large number of categories, the use of the method of ranked set sampling of several steps binary logistic regression need several steps. So it is very cumbersome to use. In view of this problem, this paper presents a new method——ranked set sampling of one—step Multiple logistics regressions , which will reduce the number of step. We use this sam-pling method for the estimation of population proportions. The numerical comparisons show that the standard deviation of the population estimates derived from the ranked set sampling of several steps binary logistic regression and the ranked set sampling of one—step Multiple logistics regressions is significantly less than simple random sampling(SRS), and the standard deviation of the population proportions estimates obtained from the ranked set sampling of one—step Multiple logistics regressions is significantly less than the ranked set sampling of several steps binary logistic regression.