"Computational Modeling of Decision-Making Tasks With a Single Line of Coding: Modeling Can Be as Easy as Doing a T-Test"
http://www.psychologicalscience.org/conventions/annual/2017-workshops
The organizers (Young Ahn and Nate Haines at Ohio State University, https://ccs-lab.github.io/) of the workshop will post the outline of the workshop and detailed instructions along with R codes & slides for the workshop here.
** Please bring your own laptop with latest R (at least 3.3.2) and RStudio installed! **
We also recommend that participants install the hBayesDM package prior to the workshop. Please click here for the instructions. Mac users, make sure Xcode is installed. Xcode is 4.5GB in size, so it might be too big to download at the workshop site.
Outline of the workshop (May 28th (Sun), 2017)
Part I (by Young Ahn) (9:00am - 9:50am)
- What is computational modeling?
- How/why do we lower the barrier to computational modeling?
- Brief introduction to hBayesDM (hierarchical Bayesian modeling of Decision-Making tasks)
- How to fit a computational model?
- Maximum likelihood estimation (MLE)
- Bayesian analysis & MCMC sampling
- Hierarchical Bayesian analysis
- Tools for Bayesian data analysis
- Things to know when performing MCMC sampling
Part II (by Nate Haines) (10:00am - 10:50am)
- Hands-on tutorial on hBayesDM (data preparatation, model fitting, model comparisons, etc.)
- Goals:
- Learn to fit models in hBayesDM
- Understand how to diagnose convergence issues when using Bayesian methods
- Learn the differences between model comparison and parameter estimation
- Have fun :D