- Treat these problems as normal problems except you have to:
- Be careful with your cross validation
best way to cross validation over time
- forward chaining cross validation is pretty popular
- hold-out cross validation also works where you only validate on the last bit of data
- although I’m afraid that you could be overfitting to the validation set (even if you run the validation multiple times with diff model seeds)
- because you aren’t looking for trends in other time periods
Link to original
- Make sure that you don’t include data from the future
- you can use the target variable (from the past) as features