Multiple Testing
posted on September 1, 2021

Multiple Testing

Multiple testing procedures are playing a key role in high dimensional problem such as variable selection and dimension reduction and genetics such as GWAS and protein domain data. One major issue in multiple testing problem is to identifying the null distribution empirically. We develop methods to estimate the null distribution based on mixture of well known distributions or empirical likelihood. method. Our current projects include some recent development of multiple testing procedures such as knockoff, mirror statistics and data splitting.

*Park, H. and Park, J. (2024) A robust false discovery rate controlling procedure using empirical likelihood with a fast algorithm. Journal of Statistical Computation and Simulation, 94(5), 1097-1120.

*Gauran, I.I., Park, J., Rattsev, I., Peterson, T.A., Kann, M.G. and Park, D. (2022) Bayesian Local False Discovery Rate for sparse count data with application to the discovery of hotspots in protein domains. Annals of Applied Statistics, 16(3), 1459-1475.

Ramos, M.L., Park, D., Lim, J., Park, J., Tran, K., Garcia, E. and Green, E. (2021) Adaptive local false discovery rate procedures for highly spiky data and their application to protein Set4Δ data, Biometrical Journal, 63(8), 1729-1744. Article Link

*Gauran, I.I., Park, J., Lim, J., Park, D., Zylstra, J., Peterson, T., Kann, M. and Spouge, J. (2018) Empirical Null Estimation using Discrete Mixture Distributions and its Application to Protein Domain Data, Biometrics, 74, 458-471. Article Link

Park, D., Park, J., *Zhong, X. and Sadelain, M. (2011). Estimation of Empirical Null Using a Mixture of Normals and Its Use in Local False Discovery Rate, Computational Statistics and Data Analysis, 55, 2421-2432. Article Link