- Course 4: Understanding a Model

Abstract: You’ve built a model!  But what is it actually doing?  Do its judgments make sense?  How do I explain it to someone else?  Why did it make this particular prediction?  This course will arm you with the knowledge and tools to address these questions.  Along the way, you will either build confidence in the fidelity and coherence of the model *or* uncover the aspect of the model that should not be trusted.

  – Module: Why should I trust a model?

  – Module: How does my model generally work?

  – Module: Why is my model making this particular prediction?

  – Module: Can I trust the probability given by my model?

  – Module: Inferring causality from models.

  – Module: Case analysis vs. population level metrics