FinCast

Financial forecasting using Machine Learning

Comment

Inspiration

Machine learning has many applications, including in the financial sector. Applications span from fraud detection to building predictive analytics tools. In this project, we focused on analytics tools for the stock market, which may be helpful in high-frequency trading.

What It Does

FinCast is a forecasting platform for stock prices and cryptocurrencies that uses Yahoo Finance data to understand trends using different mathematical and machine-learning models. We used the following:

  • ARIMA

  • LSTM

  • XGBoost

  • Hawkes processes

How We Built It

We built the different models using ARIMA and long short-term memory networks. This enables the user to make a more informed decision along with trading techniques already available.

Challenges We Faced

One of the challenges we ran into was integrating the interface with the machine-learning models and deploying the app using streamlit.

Accomplishments

We have a working prototype!

What We Learned

We have learned a lot about the different machine-learning model to forecast markets behavior.

What's Next for FinCast?

We would like to polish deploy it and then adjust the model with real-time data.

Built With

  • arima
  • google-colab
  • lstm
  • python
  • streamlit
Share this project:

Updates