Repo for the article "Multi-step time series forecasting with XGBoost"

This is the repo for the Towards Data Science article titled "Multi-step time series forecasting with XGBoost"

The article shows how to use an XGBoost model wrapped in sklearn's MultiOutputRegressor to produce forecasts on a forecast horizon larger than 1.

The approach shown in the article generally follows the approach described in the paper "Do we really need deep learning models for time series forecasting?".

The main code is found in the notebook. Some util functions are implemented in utils.py.