Riskalytics

Inspiration

As we see a lot of financial bodies rely on data to make its decision but struggles to maintain the visibility into how the data is created accessed and used. For the reason we wanted to design a project that gives the data observability through its life cycle. SO we tried to combine real time data, ML for the predictive analysis and built riskalytics.

What it does

It fetches real time stock data and gives the metrics like moving averages and volatility

How we built it

We integrated Alpha vantages, fred APIs for macro economic indicators. We processed the data and made it clean and for ML, we trained a Random Forest regressive to predict risk scores and used streamlit for the frontend development where user can upload data set to view predictions

Challenges we ran into

We encountered alpha vantage API rate limits. Handling data was complex and training the data was challenging

Accomplishments that we're proud of

Successfully built end -to-end system that integrates real time data with ML predictions

What we learned

The importance of data observability, leveraging tools like streamlit

What's next for Riskalytics

Adding more API to project to have richer analysis and then adding security features

Built With

• json
• jupyter
• matplot
• pandas
• python
• scikit-learn
• streamlit