Skip to content

jmoh3/datathon

Repository files navigation

2019 Illini Statistics Datathon

Description

In the Illini Statistics Club's 2nd annual Datathon, participants will work in teams and be presented a large, complex dataset. They will have 36 hours to apply their quantitative skills to analyze the dataset and tackle an industry problem. Though the competition is the core event of the Datathon, there will also be fun workshops presented from professionals in the field and a chance to talk to corporate sponsors. So come learn how to apply analytical skills learned in the classroom to real-life data!

Problem Statement

“Predict” the daily stock price for Bayer, Honeywell, 3M and Synchrony from the beginning of 2019 to present using quantitative and qualitative data sources.

Target Variable

Daily closing prices

Our Approach

We trained an LSTM and a Random Forest Regressor for each of the four companies whose stock prices we were trying to predict.

Our Presentation

Our Business Thesis

stonks

Libraries and APIs

  • Keras (with TensorFlow backend)
  • Sklearn
  • NLTK
  • Scrapy
  • PushShiftAPI
  • TwitterSearch
  • Plotly

About

A repository for Datathon 2019.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •