/DatascienceCompetition2018

Could we model over a billion data points (50 GB) of taxi data from Chicago to predict and interpret taxi drivers profit over the last few years? We won 2nd in the graduate division.

Primary LanguageJupyter Notebook

Taxi Profit Datascience Competition 2018

  • Final project for a competition analyzing over a billion data points (50 GB) of data.
  • We fit the data using a linear regression model with time series errors to predict median taxi revenue. 🚕
  • We won second from over 40 graduate teams!

This browser does not support PDFs. Please click this link to view it: View PDF.