First, we learned to build a model, then we performed diagnostic analysis on it. Then, we determined how accurate the model was, and in this chapter, we will extend our model-building skills. We will learn how to not view a model as an endpoint, but as a starting position to move forward toward improving models. Basically, we will learn how to improve individual models by building more than one model. We have several ways in which we can do that, and we are going to talk about them in detail.
The topics that will be covered in this chapter are as follows. These are also the ways in which we can improve individual models:
- Combining models
- Propensity scores
- Meta-level modeling
- Error modeling
- Boosting and bagging
- Continuous outcomes