Scholarlink

<p>In this article, a Bayesian alternative to the two-sample t-test and one-way ANOVA is proposed based on the Kullback-Leibler divergence and relative belief ratio. The new approach possesses attractive and desirable features since it can be calibrated, it provides evidence in favor of the null hypothesis, it allows checking the prior for prior-data conflict, and it maintains robustness in the analysis. Additionally, it does not require all populations to have a common variance. The utility of the approach is demonstrated through several examples in which it reflects its excellent performance.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

k-Sample Bayesian Test for the Mean

  • 期刊论文

  • 发表日期：

    2026年3月20日

查看全文
  • Luai Al-Labadi,
  • Ayman Alzaatreh

摘要

In this article, a Bayesian alternative to the two-sample t-test and one-way ANOVA is proposed based on the Kullback-Leibler divergence and relative belief ratio. The new approach possesses attractive and desirable features since it can be calibrated, it provides evidence in favor of the null hypothesis, it allows checking the prior for prior-data conflict, and it maintains robustness in the analysis. Additionally, it does not require all populations to have a common variance. The utility of the approach is demonstrated through several examples in which it reflects its excellent performance.