Principal investigator: James F. Troendle, Ph.D.
Research has focused primarily on simultaneous testing for differences in two groups on each of several outcome variables. Permutation and bootstrap methods for discrete outcomes have been developed that control the familywise error rate. Methods for normal outcomes have been developed to control the false discovery rate. More recently, permutation methods that control the false discovery proportion or the number of false discoveries has been developed. Current research includes methodology for testing multiple Bernoulli outcomes while adjusting for covariates and controlling the familywise error rate.
Selected Publications
Troendle JF. (2005). Multiple comparisons between two groups on multiple Bernoulli outcomes while accounting for covariates. Statistics in Medicine, 24:3581-3591. [Abstract]
Korn EL, Troendle JF, McShane LM, & Simon R. (2004). Controlling the number of false discoveries: Application to high-dimensional genomic data. Journal of Statistical Planning and Inference, 124:379-398.
Troendle JF. (2000). Stepwise normal theory multiple test procedures controlling the false discovery rate. Journal of Statistical Planning and Inference, 84:139-158.
=
