Inspiration

Not that many people are investing in stocks 39% of Americans have no money in the stock market, and over 50% say it’s rigged. 50% of black adults, and 49% of Hispanic adults don't invest, while only 32% of Asians and 28% of White Adults don’t. Moreover, less than half of young americans invest in the stock market, in fear of volatility and the 2008 market crash. Analyst target price predictions are rarely accurate.

What it does

Using data from the past, we created a ML model that predicts stock prices and more. We also incorporated this model into a website, where users can easily access predictions on all stocks. With this data, users will be able to accurately invest in stocks and attain the best profits possible. This will increase confidence in the market, and draw in more investors.

How we built it

Html5, Css3, Python were main languages Flask was chosen web framework Datetime, Pandas, and Numpy were used to process data. Scikit learn was used to generate the ML model. Yahoo Finance was used as a database Plotly was used to display graphs of data Ngrok was used to host the website. Domain.com was used to create a domain.

Challenges we ran into

We initially thought of using google cloud, but we could not connect flask to google cloud. We had a lot of trouble forwarding the domain.com domain to the ngrok, but we got that fixed. We had a hard time connecting yahoo finance with Scikit learn. We also had some complications with interpreting data with Numpy, Pandas, and datetime.

Accomplishments that we're proud of

  • We got domain.com to work with ngrok
  • We successfully made a flask website on our first attempt
  • We were able to connect the Yahoo Finance data with Scikit learn to generate a ML model.
  • We were able to display functional graphs on the website with plotly
  • We made a successful project.

What we learned

How domains work, How to create an ML model using databases like Yahoo finance, How ngrok works, How to use flask, How to use plotly to display graphs, How to use datetime, pandas, and numpy to process data.

What's next for StockWatch

  • We will add more resources to learn more about stocks
  • We will add a real time ranking system of the best stocks to buy.
  • We will incorporate data from other databases.
  • We will add more analysis features.
  • We will add a feature that predicts how much a user will gain/lose if they buy and sell a certain stock at times a user specifies.

Built With

+ 7 more
Share this project:

Updates