08-13, 12:00–12:40 (Asia/Yerevan), 213W PAB
The parameters in a statistical model are not always identified by the data. In Bayesian analysis, this
problem remains unnoticed because of prior assumptions. It is crucial to find out whether the data
determine the marginal posterior parameters. As the famous mathematician George Box stated,
“Since all models are wrong the scientist must be alert to what is importantly wrong.”
The R package ed4bhm, which is available on GitHub allows to examine the empirical determinacy of
posterior parameters for the models fitted with well-known Bayesian techniques.
- The topic (the WHAT) and WHY it is interesting:
The parameters in a statistical model are not always identified by the data. In Bayesian analysis, this
problem remains unnoticed because of prior assumptions. It is crucial to find out whether the data
determine the posterior parameters. In particular, it is important to learn to what extent the spread
and the location of marginal posterior distribution of the parameters are determined by the data.
The R package ed4bhm allows to investigate the empirical determinacy of marginal posterior
parameters, their spread and location. During this talk, I will showcase the functionality of the
package ed4bhm with an application of Bayesian logistic regression to Bacterial resistance data.
Public links to notes and helpful material will be provided on GitHub.
- The audience to WHOM the talk is addressed
This talk is interesting for everyone who is working or is interested in Bayesian analysis.
- The TAKEAWAY, a.k.a. what will I learn
The participants will learn about
1. The typical problems in a Bayesian Hierarchical Model (BHM)
2. The theory behind the empirical determinacy of posterior parameters in BHMs
3. How to fit basic BHM in R
4. How to apply the R package ed4bhm and how to interpret the results.
- Any background knowledge needed
Some basic knowledge in statistics and Bayesian statistics would be helpful.
No previous knowledge expected
Sona Hunanyan is a scientist-statistician at Philip Morris Armenia. Apart from that, she is working on research projects in statistical analysis of machine learning algorithms within the framework of the FAST Advance STEM program.
Sona completed a Master's Degree in Applied Mathematics at EPFL Lausanne. Beside her studies, she gained practical experience while working as a quantitative analyst at Nuclear Power Plant Goesgen, Switzerland. Sona holds a PhD in Biostatistics from the University of Zurich. The main topic of her PhD thesis was Bayesian hierarchical modeling and software development in R.