This is the repository for the practices of "Time series analysis", a graduate course at FaMAF's #DiploDatos, National University of Córdoba

Based in "A short course on forecasting" by Marcel Scharth, The University of Sydney.

This material draws on the Forecasting Principles and Practice textbook by Rob Hyndman and George Athanasopoulos. While the textbook and Rob Hyndman's DataCamp course use R, our tutorials provide a Python version of related content, but based only in statsmodels and the scipy stack.

This version: September 6, 2019.

• Working with time stamped data in Python
• Introduction to forecasting
• Holt-Winters smoothing
• ARIMA
• Seasonal ARIMA

Install requirements using command bellow and then open Jupyter notebook:

conda install --file conda-requirements.txt

Run and then go to http://localhost:8889

docker run -p 8889:8888 gmiretti/forecasting

To update to latest version execute the command below and run again:

docker pull gmiretti/forecasting