Virtual StatsPD@Waite meeting
- Date: Tue, 10 Aug 2021, 10:00 am - 11:00 am
- Location: Online Zoom meeting
- Contact: Beata Sznajder email@example.com
- Dr Max Moldovan Presenter
Every month, the professional development meetings of statisticians and data scientists at Waite, known as StatsPD@Waite, bring together specialists in various aspects of data sciences in agriculture from Waite, Roseworthy and Adelaide.
Please join us for the next virtual StatsPD@Waite meeting where Dr. Max Moldovan from the Biometry Hub will present on exact inference.
Please note that the StatsPD@Waite meetings are recorded. If you have a question to the speaker but would rather not be recorded, please send me your question via chat during the meeting and I will ask it on your behalf.
Please email Beata Sznajder for details of the Zoom meeting.
Exact inference revisited and the new developments in empirical data analyses: causal inference
Dr Max Moldovan (University of Adelaide Biometry Hub)
My talk will be in two parts. Firstly, I will revisit the exact inference in discrete data field, illustrating common and less common ways to exact P-values computation. Barnard's exact test will be presented as the most efficient (i.e., having maximum power, while controlling for test size, all in a frequentist inference context). Next, I will present the methods for computing confidence limits, including the method utilising the Barnard's exact testing procedure. Concerning the latter, while the resulted confidence limits may be (most) efficient, they might not be computationally feasible for large sample sizes and lack a nesting property, which is essential for applied inference.
As a second part of the talk, I will clarify and illustrate the main concepts of Mendelian Randomisation studies - the studies aiming at identification of a causal effect of an exposure on an outcome. This will bring us to the introduction of the OpenGWAS project (https://gwas.mrcieu.ac.uk) that, among other empirical analysis tools, enables implementation of Mendelian Randomisation based on 39,604 genome-wide association studies collided by the OpenGWAS project to date.