This project investigates the relationship between insider trading activity and stock price movements. By analyzing insider trading data (e.g., Form 4 filings from the SEC EDGAR database) alongside historical stock price data, the project aims to identify patterns and correlations that can help retail investors better understand how insider trades may influence stock performance. The ultimate goal is to develop a prototype system that visualizes these trends and offers insights into potential stock price impacts based on insider trading activity. More infomation for the project is mentiond in Class project-insider trades.docx
- Backend -> Python, FastAPI
- Database -> MongoDB
- Frontend -> React, Material UI
- Deployment -> AWS/GCP
This document provides a brief overview of the backend API endpoints.
- POST /auth/signup: Create a new user account.
- POST /auth/token: Log in and obtain an access token.
- GET /auth/me: Fetch current user's information (requires authentication).
- PUT /auth/me/update: Update user profile (requires authentication).
- GET /transactions/{ticker}/{transaction_id}: Fetch a specific insider trading transaction.
- GET /transactions/{ticker}: Retrieve all transactions for a given ticker.
- GET /stocks/{ticker}: Retrieve historical and current stock data.
- POST /future: Generate stock price forecasts based on input data.
- GET /: Welcome message.
Most endpoints require authentication with a valid access token in the Authorization header:
We have used the Swagger UI for API documentation and testing. The Swagger UI can be accessed at the following URL:
http://cmpe272teamsnooptrade.us-west-2.elasticbeanstalk.com/docs
This frontend provides a user-friendly interface for visualizing insider trading patterns and stock trends.
- Dashboard: Interactive graphs and tables for stock-insider trade insights.
- Authentication: Google OAuth and email/password login.
- About: Project overview and team information.
- Account: User profile and settings.
- React
- Material UI
- Axios
- Chart.js
- Communicates with the backend via REST APIs.
- Deployed on AWS/GCP for scalability.
http://cmpe272frontend.us-west-2.elasticbeanstalk.com/dashboard
https://docs.google.com/spreadsheets/d/1uJr3OGZMIMjZN1VK0RzqEzCFX1NDH_OOJ3BB7nGJSpk/edit?usp=sharing
