The Bayesian Statistical Science Section of the ASA provides a forum for statisticians and people who have interest in the Bayesian paradigm. The broad objectives of the Section are: to encourage research on theory and methods of statistical inference and decisionmaking associated with Bayes' theorem and to encourage the application and proper use of Bayesian procedures in the behavioral, biological, managerial, engineering, environmental, legal, medical, pharmaceutical, physical, and social sciences.
Upcoming Web-Based Lectures
Title: Graphical Approaches to Multiple Testing
Presenter: Frank Bretz, Novartis Pharma AG, and Dong Xi, Novartis Pharmaceutical Company
Date and Time: Thursday, March 5, 2015, 10:00 a.m. - 12:00 p.m. Eastern time
Sponsor: Biopharmaceutical Section
Registration Deadline: Tuesday, March 3, at 12:00 p.m. Eastern time
Methods for addressing multiplicity are becoming increasingly more important in clinical trials and other applications. In the recent past, several multiple test procedures have been developed that allow one to map the relative importance of different study objectives as well as their relation onto an appropriately tailored multiple test procedure, such as fixed-sequence, fallback, and gatekeeping procedures. In this webinar we focus on graphical approaches that can be applied to common multiple test problems, such as comparing several treatments with a control, assessing the benefit of a new treatment for more than one outcome variable, and combined non-inferiority and superiority testing. Using graphical approaches, one can easily construct and explore different test strategies and thus tailor the test procedure to the given study objectives. The resulting multiple test procedures are represented by directed, weighted graphs, where each node corresponds to an elementary hypothesis, together with a simple algorithm to generate such graphs while sequentially testing the individual hypotheses. We also present several case studies to illustrate how the approach can be used in practice. In addition, we briefly consider power and sample size consideration to optimize a multiple test procedure for given study objectives. The presented methods will be illustrated using the graphical user interface from the gMCP package in R, which is freely available on CRAN.
Title: Design, Weighting and Variance Estimation for Population-based Evaluation Studies
Presenter: David Judkins, Principal Scientist, Abt Associates
Date and Time: Thursday, March 19th, 12:30 p.m. - 2:30 p.m. Eastern time
Sponsor: Survey Research Methods Section
Registration Deadline: Tuesday, March 17, at 12:00 p.m. Eastern time
Most years, there are a few really large population-based evaluation studies going on of federal programs designed to improve the economic well-being and health of disadvantaged domestic populations. They are typically sponsored by evaluation divisions of the Departments of Labor, Agriculture, Education, and Health and Human Services. One of the largest in U.S. history is now being conducted by the Social Security Administration on ways of encouraging disabled adults to return to the labor force. These evaluations often involve true experimental designs, but may also involve quasi-experimental designs and regression discontinuity designs. Sometimes the studies rely on only either administrative data or follow-up survey data to measure outcomes, but often both follow-up survey and administrative data are used to measure outcomes. Usually some degree of clustering is employed in the design - possibly to make collection of outcome data more efficient, but more often because of resource constraints for treatment delivery or monitoring of treatment delivery. Probabilities of treatment assignment often drift over time in response to local treatment capacities. The combination of clustering, differential treatment assignment probabilities, follow-up survey nonresponse, and linked administrative data make for an interesting set of challenges very similar to those encountered in the design and analysis of descriptive population surveys. In addition, if only administrative data are used, sample sizes can be very large, approaching survey sample sizes otherwise seen only in the American Community Survey. These sample sizes imply data processing challenges for resampling-based variance estimation and multiple-comparison adjustment procedures. This course will present solutions to many of these more interesting challenges that are aligned with survey methods issues.
Title: Statistical Methods Used in Pre-Clinical Drug Combination Studies
Presenter: Wei Zhao, Medimmune
Date and Time: Thursday, April 16, 2015, 12:00 p.m. - 2:00 p.m. Eastern time
Sponsor: Biopharmaceutical Section
Registration Deadline: Tuesday, April 14, at 12:00 p.m. Eastern time
Various oncogenic cell signaling pathways are known to provide cross-talk and redundancy within tumors. Thus inhibition of such pathways individually by a single targeted therapy has been shown to lead to compensation by other pathways. This, in turn, results in a loss of sensitivity to the original targeted therapeutic agent at the cellular level. In the clinic, this type of compensation leads to tumor resistance and relapse. Because advanced tumors are often resistant to single agents, there is an increasing trend to combine drugs to achieve better treatment effect and reduce safety issues. The growing interest in using combination drugs has spawned the development of many novel statistical methodologies. In this webinar presentation, I will demonstrate the various statistical methods used in designing and analyzing pre-clinical drug combination studies.
*** More information and registration at the webinar page.