Abstract

The comparison of the means of two treatments or populations when more than one variable is measured may be done using Hotelling’s T2 statistic. In many real world situations the data obtained are mixed, i.e. one variable is dichotomous and the other variable is continuous. The assumption of multivariate normality upon which Hotelling’s T2 is based is no longer valid. In this paper, an approximate Hotelling T2 test is proposed for bivariate mixed data and empirically evaluated in terms of Type I error rate. It is shown that the approximation does a good job of controlling the Type I error rate for a range of bivariate mixed parameters even for relatively small sample sizes.

Keywords

bivariate mixed data, dichotomous response, Hotelling T2, multivariate analysis

Creative Commons License


This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Share

COinS
 
Apr 25th, 5:00 PM

HOTELLING’S T2 APPROXIMATION FOR BIVARIATE MIXED (DICHOTOMOUS & CONTINUOUS) DATA

The comparison of the means of two treatments or populations when more than one variable is measured may be done using Hotelling’s T2 statistic. In many real world situations the data obtained are mixed, i.e. one variable is dichotomous and the other variable is continuous. The assumption of multivariate normality upon which Hotelling’s T2 is based is no longer valid. In this paper, an approximate Hotelling T2 test is proposed for bivariate mixed data and empirically evaluated in terms of Type I error rate. It is shown that the approximation does a good job of controlling the Type I error rate for a range of bivariate mixed parameters even for relatively small sample sizes.