| Name | Student ID |
|---|---|
| Chetan Dhingra | 1862481 |
| Hish Salehi | 1812352 |
| Velda Iskandar | 1882870 |
How does news sentiment affect stock prices?
This project aims to understand if the general feeling (sentiment) expressed in news articles about a company can help predict how its stock price will change. We will collect news articles, analyze their sentiment, and then see if this sentiment is related to actual stock price movements. We will use machine learning techniques to build a model that tries to predict these price changes based on news sentiment.
- Get sentiment scores from news articles about a specific stock.
- See if there's a connection between these sentiment scores and how the stock price changes.
- Build computer models to predict stock price changes using news sentiment.
- Check how well our models perform using standard statistical measures.
- Use methods like MCMC sampling to better understand our model.
We will use financial news articles and historical stock price data. The specific sources and how often we get the data will be decided later. We will choose data that allows us to see both quick reactions to news and longer-term trends.
We will first process the news articles to determine their sentiment (positive, negative, or neutral). Then, we will use this sentiment data along with historical stock prices to build our prediction models. We plan to use Bayesian inference techniques for this.
We will measure how well our models work by looking at metrics like AUC-ROC. This will tell us how good our model is at correctly predicting whether a stock price will go up or down.
| Member | Role |
|---|---|
| Chetan Dhingra | Data Collection and Preprocessing |
| Hish Salehi | Model Modeling and Implementation |
| Velda Iskandar | Evaluation and Report Writing |
This project will help us understand the relationship between news and stock markets. It could provide insights for investors on how news sentiment might influence stock prices.