ATOM is meeting on Tuesday, July 16th, 6:30pm, at Galvanize!
This month, Erik Case will be leading a discussion on time-series forecasting : a look back and ahead.
Once upon a time in the 1970s, the Box-Jenkins approach to forecasting time series landed like a thunderclap, displacing aged and brittle econometric forecasting models with a systematic approach ARIMA modeling. While predicting the movement of sequential data is a hugely important academic/business problem, time series forecasting remains a niche in applied statistics and ML.
The classic ARIMA approach is elegant, but requires considerable specialization to tune properly. We will discuss a brief history of applied ARIMA forecasting and close out with a case study using a modern, alternative GAM-like forecasting approach found in Facebook’s open source Prophet package. This approach is more user friendly, but also requires more domain knowledge of the time series being forecasted.
The reading for this session is Forecasting at Scale (Prophet) https://peerj.com/preprints/3190.pdf
If you are interested in diving into the source code, this library is maintained at https://github.com/facebook/prophet.
If you would like suggested background in classical time series modeling : Forecasting: Principles and Practice (Chapter 8 - ARIMA) https://otexts.com/fpp2/arima.html
Advanced Topics on Machine learning ( ATOM ) is a learning and discussion group for cutting-edge machine learning techniques in the real world. We work through winning Kaggle competition entries or real-world ML projects, learning from those who have successfully applied sophisticated data science pipelines to complex problems.
As a discussion group, we strongly encourage participation, so be sure to read up about the topic of conversation beforehand !
ATOM can be found on PuPPy’s Slack under the channel #atom, and on PuPPy’s Meetup.com events.
We're kindly hosted by Galvanize (https://www.galvanize.com). Thank you !