Presented by:Nicola Justice, Robin Lock, Allan Rossman, and Chris Wild
This panel will begin by discussing WHY teach simulation-based inference, and then move into four different perspectives on HOW to teach simulation-based inference, addressing several choices to be made. In addition to questions from the audience, the following questions will be addressed:
- Why teach simulation-based inference?
- What order do you teach topics in? Do you think order matters? Where do traditional (normal and t-based) methods get incorporated, if at all?
- How do you teach interval estimation, and why do you teach it this way?
- Two general types of simulation (“cranks”) exist: the reallocating/shuffling/scrambling crank and the resampling crank. Should students see both types of cranks?
- What features do you like in software for teaching simulation-based inference? What should the students have to do and what should the software do for them? What role does animation play? What are the biggest conceptual-connections to be addressed with software? How does software for developing conceptual understanding differ from software for doing data analysis?
(Tip: click the fullscreen control)
Having trouble viewing? Try: Download (.mp4)
(Tip: right-click and choose "Save As...")Teaching with Simulation-Based Inference