You can install the released version of StatsForecast from the Python package index with:

pip install statsforecast


conda install -c conda-forge statsforecast


We are constantly updating StatsForecast, so we suggest fixing the version to avoid issues. pip install statsforecast=="1.0.0"


We recommend installing your libraries inside a python virtual or conda environment.

User our env (optional)

If you don’t have a Conda environment and need tools like Numba, Pandas, NumPy, Jupyter, StatsModels, and Nbdev you can use ours by following these steps:

  1. Clone the StatsForecast repo:
$ git clone && cd statsforecast
  1. Create the environment using the environment.yml file:
$ conda env create -f environment.yml
  1. Activate the environment:
$ conda activate statsforecast


The following features can also be installed by specifying the extra inside the install command, e.g. pip install 'statsforecast[extra1,extra2]'

  • polars: provide polars dataframes to StatsForecast.
  • plotly: use StatsForecast.plot with the plotly backend.
  • dask: perform distributed forecasting with dask.
  • spark: perform distributed forecasting with spark.
  • ray: perform distributed forecasting with ray.