In this module, we will discuss how one can assess the quality of different models, and what aspects of model quality one needs to consider.

Learning Objectives

  • Learn how to determine different aspects of model quality
  • Become familiar with different types of model performance metrics
  • Become able to choose the right assessment metric for your problem
  • Learn what model uncertainty is and how to compute it
  • Become familiar with model diagnostics


If we fit any kind of statistical model to data, we need to determine the quality of a fit somehow. How do we know one model is better than another? How do we know any given model is good? These questions, i.e., what model is better among a group of models, and if a model is good (even the best among a group of awful models can be a bad model!) have multiple parts to them, only some of which can be answered by statistics. Others are judgment calls.

There is no single way to define quality. Instead, multiple factors should be considered. Here are some important aspects to think about:

  • Model performance: How well do the predictions from the model compare to the actual data?
  • Model complexity: How easy or hard is it to understand the model? How fast does it run? How robust is it to violations of the assumptions you made?
  • Model uncertainty: How much uncertainty is associated with the predictions made by the model?
  • Model misspecification: How much systematic deviation is there between model predictions and actual data?

We will discuss these different ways one can assess model quality in this module.