Skip to content

Used time series modeling of 750K relevant 311 requests to predict the volume of weekly rat activity for every NYC neighborhood, with an average MAPE of 5.6%, which currently feeds into a live PowerBI dashboard which informs targeted rat mitigation efforts

Notifications You must be signed in to change notification settings

claireboyd/predicting_rats_nyc

Repository files navigation

Time-Series Forecasting with Prophet

This repository is for a hands-on presentation focused on using prophet for time series analysis. It covers:

  • a high-level overview of time series analysis
  • an illustration of Facebook's python package called prophet by using AirPassengers data, modeling the number of passengers flying to Austrailia in the 1960s.
  • an application of prophet in the public sector (predicting rat activity in NYC neighborhoods)

For more information about this work and a step-by-step guide, see the ipynb file prophet_session.ipynb. Thanks!

If you'd like to learn more about my approach or have questions on the methods, please feel free to reach out to me directly at [email protected].

Implementation notes:

If you want to convert / update the .ipynb into .slides.html, you can do so using the jupyter nbconvert module, issuing the following command within your project directory:

jupyter nbconvert prophet_session.ipynb --to slides --post serve

About

Used time series modeling of 750K relevant 311 requests to predict the volume of weekly rat activity for every NYC neighborhood, with an average MAPE of 5.6%, which currently feeds into a live PowerBI dashboard which informs targeted rat mitigation efforts

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published