ON ENHANCED ANALYSIS OF MEANS USING COVARIATES

In this era of Industry 4.0, the digitalization of processes creates opportunity to store many quality characteristics. Most industrial researchers are mainly interested in examining the equivalence of treatment means. The analysis of means (ANOM) is the most proper graphical test used to compare the treatment means. However, in real life situations, some linearly related covariates are also recorded along with the main variable. In such cases, ANOM does not consider the effect of the covariate while checking the equivalence of means. The analysis of means for covariate (ANOMC) has recently been proposed based on the well-known regression estimator (say ANOMC-Reg). It is observed that ANOMC-Reg has poor performance under the extremely positive skewed distribution; hence, this study is intended to improve the performance of the ANOMC test using ratio estimator, and the proposed test is referred to as the ANOMC-R test. Finally, a simulation study is designed, and the performance of both tests is compared under different design parameters using the tests' percentage type I error and power. The findings reveal that the proposed test is more efficient than the existing one under the said positively skewed environments.