Minnow Analytics

Inspiration

The basis for our project is that we don't see many unbiased stock recommendations from financial and wealth management experts. Advisors are inclined to pick certain stocks over others, which can easily trick and manipulate their money into stocks, especially when they don't know much about finance.

What it does

Our model uses machine learning to take in past prices and predict the future price of the stock with what it has learned from the past

How we built it

We first used a simple EMA calculation to gather a window of data and calculate the stock price. We then used the machine learning model to calculate the price from windows. We then connect MongoDB database to Yahoo Finance to get the time and price of the stock you want as the data. We then implemented this onto our website which is simple and easy to use for all users.

Challenges we ran into

One challenge we ran into was data framing and adding dates into data frames. This proved to be very complex as we had trouble adding the correct time to the dates. Another problem that we faced was implementing the EMA algorithm as well as understanding how the machine learning model we implemented worked.

Accomplishments that we're proud of

We are proud of machine learning and connecting a MongoDB database to our backend systems as none of us formally learned how to do that before, we also developed our skills in machine learning and the many packages on Python we can utilize. We also gained insight into the streamlit platform which is something we are very proud of as none of us had used the product before.

What we learned

We learned about how to utilize APIs, connect databases to data sets, as well as utilize machine learning models to predict time series algorithms.

What's next for Minnow Analytics

We plan on adding more factors into our calculations such as past interest rates, comparable companies, cash flow analysis, and EPS to make a better prediction for our stocks.

Built With

• api
• ema
• keras
• machine-learning
• mongodb
• web