Predicting churn
In this exercise, you will use the predict()
function in the pROC
package to predict the churn probability of the customers in the test set, test_set
.
The function has three arguments:
- The model used to make the predictions.
newdata
: The dataset the model should be applied to.type
: the type of prediction. We want to know the probability of churn, so we choose eitherresponse
orprob
.
Use the function predict()
to predict the churn probability of the customers in test_set
.
This exercise is part of the course
Predictive Analytics using Networked Data in R
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Load the package
library(pROC)
# Predict with the first model
firstPredictions <- predict(___, newdata = test_set, type = "response")