Working with time-series. the aim of the project is to predict sales by items and stores. Due to the complexity of the problem I decided to turn it into a regression problem, but the problem needs to remain relative to the time, so, advanced feature engineering was applied, together with the ensemble stacking model.
Stacking
1st models:
- Catboost
- XGBM
- Random forest
- Linear Regression
- KNN
2st model:
- Linear Regression