Single-stage sampling procedure for heteroscedasticity in multiple comparisons with a control

Abstract

Given k (≥2) independent and normally distributed populations with unknown means and unknown (and possibly unequal) variances, we used the single-stage sampling procedure for multiple comparison with a control to the problem of unequal sample sizes. A simulation of the family-wise error type I was conducted to validate the quality of the procedure and a numerical example was provided for illustrating this procedure. We provided a user-friendly interface by R shiny to obtain the critical values for the practitioners.

Keywords:

• Simultaneous confidence intervals
• Student t distribution
• Unknown and unequal variances

Conflict of interest statement

The authors declare no conflicts of interest.

Data availability statement

All authors make sure that all data support their published claims and comply with field standards.

Additional information

Funding

Wen’s and Wang’s research was sponsored by the Ministry of Science and Technology, Taiwan (MOST 109-2118-M-006-004-MY2).