Learning objectives

1. Statistical theorems prefer larger sample, and indicate that they will not inflate false positive rates, but why are there so many complains from applied researchers?
2. How reliable are the p-values provided by current differential expression analysis of large sample genomic data?
3. Our solution: distribution free and dynamic model complexity.

Speaker bio:

Dr. Xuekui Zhang is a Canada Research Chair (Tier 2) and a Michael Smith Health Research BC Scholar. Dr. Zhang focuses on developing and applying novel statistics methods and software tools to unravel the intricacies of big data in a wide array of domains, including medical research, agricultural sciences, economics, and the study of environmental ecosystems. His research interests include (1) Bioinformatics, (2) Machine learning, (3) Design of clinical trials, and (4) COPD.