Abstract

Covariates are important aspects of models, as well as identifying causes and effects within the system. It is useful to think about big picture for the model to identify variables that may be important. Reviewing the data in different visualizations with potential variables, such as Snow’s map, can be helpful. After variables are identified, the next step is to decide or test what types of covariates they may be. When covariates are important to include, their statistical terms will be significant. Add to that the model’s overall fit by R**2 or a reduction in the standard error of the estimate, and the prediction power improves.