Hi All,
I would like to predict in which month the individual customer is more likely to transact or to miss transactions. I have 2 years data (I have attached a sample).
What is the right technique to use in this case?
In case you can't see the attached spreadsheet, please see below how the data looks like..
Thank you so much for your help
customer_id | WEEK_START_DATE | WEEK_END_DATE | No_trans | Tot_sales |
50026 | 30-Aug-15 | 05-Sep-15 | . | . |
50026 | 06-Sep-15 | 12-Sep-15 | . | . |
50026 | 13-Sep-15 | 19-Sep-15 | . | . |
50026 | 20-Sep-15 | 26-Sep-15 | . | . |
50026 | 27-Sep-15 | 03-Oct-15 | . | . |
50026 | 04-Oct-15 | 10-Oct-15 | . | . |
50026 | 11-Oct-15 | 17-Oct-15 | . | . |
50026 | 18-Oct-15 | 24-Oct-15 | 2 | 4 |
50026 | 01-Nov-15 | 07-Nov-15 | 1 | 2 |
50026 | 08-Nov-15 | 14-Nov-15 | 2 | 4 |
50026 | 15-Nov-15 | 21-Nov-15 | 3 | 6 |
50026 | 22-Nov-15 | 28-Nov-15 | . | . |
50026 | 29-Nov-15 | 05-Dec-15 | . | . |
50026 | 06-Dec-15 | 12-Dec-15 | . | . |
50026 | 13-Dec-15 | 19-Dec-15 | 1 | 2 |
50026 | 20-Dec-15 | 26-Dec-15 | . | . |
Poisson Regression Model ?
Check
1) PROC GENMOD + REPEATED
2)PROC GLIMMIX
Available on demand!
Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.
Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.
Find more tutorials on the SAS Users YouTube channel.