Practical Predictive Analytics

Using formula notation in an rpart model

We have already seen how to specify regression models in R using formula notation. For example, we use lm(women$height ~ women$weight)) to predict height based upon weight.

The general form of formula notation in regression models is:

Dependent variable ~ Independent variables

Where the dependent variable is on the left side of the tilde (~), and the independent variables are on the right side of the tilde.

An rpart model can be set up using formula notation as well, with a slight change in terminology. In an rpart model, we can specify the target variable (Life.Exp) on the left side of the tilde (~), and the predictor variables on the right side of the tilde. We will use dot notation (.) to designate ...