Model-Based Clustering for High Dimensional Data
Shahina Rahman
Department of Statistics, Texas A&M University
Valen E. Johnson
Department of Statistics, Texas A&M University
Irina Gaynanova
Department of Statistics, Texas A&M University
Anirban Bhattacharya
Department of Statistics, Texas A&M University
The talks in this session will be focused on the recent debates surrounding the use/abuse/misuse of p-values and issues of reproducibility in science. The topic is timely, due to the recent ASA Statement on p-values, as well as moves within the scientific communities to bring more clarity and transparency to the reporting of statistical analysis and results. The session will host four invited speakers, as follows (name, affiliation, contact information and tentative title are provided): * Naomi Altman, Penn State University (nsa1@psu.edu) P-values, power and reproducibility - approaches from high-throughput biology * Andrew Gelman, Columbia University (gelman@stat.columbia.edu) Resolving the reproducibility crisis using Bayesian inference * Valen Johnson, Texas A&M University (vjohnson@stat.tamu.edu) Comments on marginally significant p-values * Jeffrey Leek, Johns Hopkins University (jtleek@jhu.edu) Reproducibility is solved, p-values aren't the problem, and its time for the real work of data science