(c) 2016 Justin Bois. This work is licensed under a Creative Commons Attribution License CC-BY 4.0. All code contained herein is licensed under an MIT license.
This tutorial exercise was generated from an Jupyter notebook. You can download the notebook here. You can also view it here. Use this downloaded Jupyter notebook to fill out your responses.
Describe three ways of reporting numerical summaries of the posterior (error bars), and their relative merits.
Give an example of a situation where it is ok to throw out data you acquired.
What do we mean by "heavy tail?" Why are heavy-tailed distributions useful when dealing with potential outliers?
If you were to summarize the posterior with a MAP/error bar using optimization (like we did in week 3, not MCMC), would it be easier to use the Cauchy method of dealing with outliers or the good/bad data model? Why?