Application of kappa statistics in sequential tests for family-based design

Authors

FARID RAJABLI

DOI

10.3906/elk-1401-283

Abstract

Family-based designs are commonly used in genetic association studies to locate markers associated with diseases. It is a challenging task to collect a large enough sample size, perform a statistical test, and obtain the desired statistical power. The sequential probability ratio test (SPRT) was introduced to overcome the limited sample size problem. However, the drawback of SPRT is that, for the sake of accuracy, the test leaves many markers in a gray zone meaning ``no decision''. In this article, we propose a novel approach: a sequential probability ratio test `plus' (SPRT$+)$ to reduce the number of these gray zone markers. Using simulated data, the results of SPRT$+$ are compared with the results of SPRT. SPRT$+$ shows a promising overall performance in identifying highly and moderately associated markers in the correct association region without a loss of accuracy.

Keywords

Markers, sequential probability ratio test, simulation study, family-based association study, kappa statistics

First Page

1984

Last Page

1991

Recommended Citation

RAJABLI, FARID (2016) "Application of kappa statistics in sequential tests for family-based design," Turkish Journal of Electrical Engineering and Computer Sciences: Vol. 24: No. 3, Article 90. https://doi.org/10.3906/elk-1401-283
Available at: https://journals.tubitak.gov.tr/elektrik/vol24/iss3/90