CryptoFrontier

• 

  Output graph

• 

  Input interface

• 

  Suggestions

• 

  Title

Inspiration

Personal interest in the blockchain technology together with the realisation that a lot of new investors do not use solid optimisation tools for their crypto portfolios, potentially taking too much risk and losing out on potential returns.

What it does

Applies efficient frontier theory to cryptocurrency portfolios. It takes as an input the users portfolio and pits it against randomly simulated portfolios along this efficient frontier curve. It uses historic data from a cryptocurrency exchange and suggests what changes the user should make to their portfolio to maximise their expected gains while minimising their expected losses.

How we built it

The code was written in Python using the streamlit framework. The backend leverages numpy and pandas for data science and rapid matrix computation.

Challenges we ran into

Pulling accurate data from the online databases, interpreting high dimensional mathematical formulas, and building a user friendly frontend.

Accomplishments that we're proud of

We have built a full functioning web-app which does exactly what we wanted it to do. The code is fully open source and a live web app is available.

What we learned

Financial concepts, web-app development steps, frontend (streamlit), matrix compuation (backend).

What's next for CryptoFrontier

RIght now it runs on the streamlit framework, ideally it would become a stand alone web-app or an app for smartphones that would allow user sign-up and additional features. Optimise the code to run more efficiently. Add more features such as suggesting which cryptocurrencies to take out of the portfolio and which the user might want to add to diversify their portfolio.

Built With

• python
• streamlit