I have a dataset with some missing data. The missing data is categorical and represented by bins (please, see example below: 'purchase_price', 'customer_income', etc.). What is the best approach for imputing data like this? Should I transform the bins first? Somehow, cannot find any recommendations online.
| purchase_price | trade_in | vehicle_finacing | customer_age | customer_income |
|---|---|---|---|---|
| 15001-20000 | 1 | 1 | 21 - 30 | 40001-60000 |
| 15001-20000 | 0 | 0 | 51-60 | 0-20000 |
| 25001 - 30000 | 1 | 1 | 41-50 | 60001-80000 |
| 10001 - 15000 | 0 | 1 | 21-30 | 60001-80000 |
| 25001 - 30000 | 1 | 1 | 31-40 | 120000-140000 |