Adaptive Group-combined P-values Test for Two-sample Location Problem with Applications to Microarray Data.

Area of Science:

• Statistics
• Biostatistics
• High-dimensional data analysis

Background:

• High-dimensional data analysis presents challenges for traditional statistical tests.
• Existing methods often fail when data dimension exceeds sample size or lacks normality.
• Type I error rates can be compromised in high-dimensional settings.

Purpose of the Study:

• To propose a novel statistical test for the two-sample location problem in high-dimensional data.
• To develop a test robust to high dimensionality and non-normal distributions.
• To address limitations of existing tests in challenging data scenarios.

Main Methods:

• An adaptive group p-values combination test is proposed.
• The method combines p-values adaptively to enhance robustness.
• The test is designed to be effective regardless of data distribution or dimension-sample size relationship.

Main Results:

• Simulation studies demonstrate correct control of Type I error rates.
• The proposed test shows superior performance compared to existing methods in most scenarios.
• The method was successfully applied to Ageing Human Brain Microarray data.

Conclusions:

• The adaptive group p-values combination test is a powerful and robust tool for high-dimensional two-sample location problems.
• This approach offers a reliable solution for complex biological and statistical data.
• The method provides accurate error control and enhanced statistical power.