I have tried to create a simple case where ColumnarModelData should be able to find the pattern but it doesn’t work.
I have created a dataset where y is the sales and the data also includes unit price and quantity.
It should be able to easily find qty * price = sales but it doesn’t.
I have tried unit price and qty bot as category variable and continuous but neither seem to work
Have I misunderstood how this is meant to work?
The data is below.
| Date | Product | unit price | quatity | Sales |
|---|---|---|---|---|
| 2014/5/1 | P1 | 100 | 53 | 5300 |
| 2015/5/1 | P1 | 100 | 2 | 200 |
| 2016/1/1 | P1 | 100 | 88 | 8800 |
| 2016/3/1 | P1 | 100 | 108 | 10800 |
| 2016/4/1 | P1 | 100 | 30 | 3000 |
| 2016/6/1 | P1 | 100 | 38 | 3800 |
| 2016/9/1 | P1 | 100 | 2 | 200 |
| 2016/12/1 | P1 | 100 | 64 | 6400 |
| 2017/3/1 | P1 | 100 | 60 | 6000 |
| 2017/5/1 | P1 | 100 | 21 | 2100 |
| 2017/8/1 | P1 | 100 | 33 | 3300 |
| 2017/10/1 | P1 | 100 | 76 | 7600 |
| 2017/12/1 | P1 | 100 | 58 | 5800 |
| 2018/1/1 | P1 | 100 | 22 | 2200 |
| 2018/2/1 | P1 | 100 | 108 | 10800 |
| 2018/4/1 | P1 | 100 | 89 | 8900 |