MJE Stock Evaluator

• 

  The Main Homepage

• 

  An example report using Apple's Stock

Inspiration

Going into this weekend, our group had a strong interest in stocks. Our goal was particularly to find how AI may be able to fit into the stock trading process.

What it does

MJE Stock Evaluator allows a user to input a stock of their choice and observe a data-backed analysis of whether they should buy, hold, or sell a stock.

How we built it

We built the evaluator using a python script for the backend analysis and summarization which utilized python libraries including yfinance (for data), numpy (for basic data analysis), and Technical Analysis (for more complex data analysis).

Challenges we ran into

The largest challenge we ran into was the analysis producing an identical result for several stocks of similar performance. This was due to a small amount of parameters being place into the formula which decided the recommended action for a stock. To fix this, we altered the decision making portion of our program from a hard-coded program to a model-based system which gave more variance in its results.

Accomplishments that we're proud of

We are proud of creating an MVP for this product given that none of us had any prior experience in several of the tools used including quantitative analysis, React, or OpenAI.

What we learned

Similar to above, we learned several new formulas used for quantitative analysis including SMA-20, SMA-50, and RSI. Additionally, we learned the fundamentals of React and how to use OpenAI developer tools.

What's next for MJE Stock Evaluator

Next, we plan to add a way in which users can set a stock to be automatically evalutated at some periodic unit of time, such as every hour. Then, the script can notify the user of whenever the script changes from 'hold' to either 'buy' or 'sell.'

Built With

• numpy
• openai
• python
• react
• ta
• yfinance