Testing for a General Changepoint in Medical and Psychometric Studies: Changes Detection and Sample Size Planning

Area of Science:

• Statistical methods
• Biostatistics
• Psychometrics

Background:

• Change detection is crucial in medical and psychometric research.
• Existing methods for change point detection can be computationally intensive, requiring resampling or simulations.
• Power analysis and sample size planning are essential for study design.

Purpose of the Study:

• Introduce a novel, computationally simple method for change detection using the pseudo Score statistic.
• Determine the sampling distribution of the pseudo Score statistic under the alternative hypothesis.
• Facilitate power level and sample size calculations through known null and alternative distributions.

Main Methods:

• Utilized the pseudo Score statistic for change detection.
• Determined the sampling distribution under the alternative hypothesis.
• Conducted simulation studies with normally distributed and binary data.
• Performed real data analyses on genomic and SAT critical reading datasets.

Main Results:

• The proposed method demonstrated superior performance compared to existing Tests for a Change Point (TFCP).
• The method simplifies power analysis and sample size estimation in segmented regression.
• The approach eliminates the need for computationally expensive resampling or simulations.

Conclusions:

• The new pseudo Score statistic method offers a significant advancement in change detection for medical and psychometric studies.
• The method's computational simplicity and known distributions enhance its utility for power and sample size planning.
• The availability of the method on CRAN and as a Shiny App promotes its adoption in research.