top of page

TEAM

Analyzing the Impact of News Topics on Stock Prices

James Steele, Enrico Antonio La Vina, Xiangwei Peng, Xiaokang Wang, Joel Schargorodsky

clear.png

Our project aims to investigate the relationship between news topics and stock price fluctuations, focusing on how various news categories, beyond just financial news, affect specific companies' stock performance.
Our rough idea is to start by using clustering to group news topics without predefined categories, followed by identifying which clusters most strongly correlate with stock price changes. Using these high-impact clusters and sentiment analysis, we'll train LSTM models to predict stock price movements. Though we don’t have any data sets finalized at this time, we aim to identify ones that include comprehensive news data, financial records, and potentially real-time news APIs. Here are a few contenders:
- https://www.kaggle.com/datasets/rmisra/news-category-dataset 
- https://components.one/datasets/all-the-news-articles-dataset/ 

Screen Shot 2022-06-03 at 11.31.35 AM.png
github URL
bottom of page