Track 3: Inference
Track 3: Inference
Overview
Advanced statistical methods for parameter estimation and model calibration.
Prerequisites
- Any form of model fitting experience in which you tweak a model to better align with data (even if in casual or nonstandard ways)
- At least know the concept of “likelihood”
Course Content
This track focuses on model fitting / parameter estimation approaches: once you have your model and data, how you should adjust the parameters so that the model best represent your data. Topics to be covered include maximum likelihood estimation, Bayesian inference / MCMC, model assessment and selection, etc. Emphasis will be placed on the theoretical backgrounds, rationales and principles rather than on specific techniques (althoug some will be covered). The goal is to make sure that you know what you are doing, including its limitations and pitfalls, instead of blindly using exsiting tools.
“If you think you don’t know about inference, come to this track. If you think you know about inference, more reasons for you to come to this track”