For this exercise, I've created an artificial dataset; to follow along, you can download it from GitHub:
https://github.com/datameister66/data/blob/master/sequential.csv
There are also datasets available with the package and tutorials are available. My intent was to create something new that mirrored situations I have encountered. I developed it completely from random (with some supervision), so it does not match any real world data. It consists of 5,000 observations, with each observation, the history of a customer and nine variables:
- Cust_segment--a factor variable indicating the customer's assigned segment (see Chapter 8, Cluster Analysis)
- Eight discrete purchase events named Purchase1 through Purchase8
