Subsetting
Remember, the company wants to expand their business and offer Hoppiness nationwide. Therefore, the model should also work on data obtained from new stores. One possibility to check how your model will perform on new data is to first train your model on one part of the data and then try to predict the remaining part of the data.
You create a trainings data set by withholding the last purchase that was recorded for each customer. You can do this by using the function subset()
on choice.data
and selecting all observations with indicator variable LASTPURCHASE
to equal 0
. Likewise, you create a test data set by selecting all observations with indicator variable LASTPURCHASE
to equal 1
.
This exercise is part of the course
Building Response Models in R
Exercise instructions
- Leave out the last purchase to create a training data set. Use the function
subset()
onchoice.data
. SpecifyLASTPURCHASE == 0
assubset
argument. Assign the result totrain.data
. - Create a test data set by using the function
subset()
onchoice.data
. SpecifyLASTPURCHASE == 1
assubset
argument. Assign the result totest.data
.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Create the training data
# Create the test data