Inspiration

A project involving trading is a perfect fit for us, since we have members whit an economics background and multiple members of our team own stocks. We know first hand how difficult it can be to make sense of large amounts of data (e.g. SEC filings). This project gave us the opportunity to combine our passion for technology and data analysis with our interest in the financial world.

What it does

Our web application allows traders to make technical analyses and gauge the general sentiment of the market through real time Forexnews. We display Forex exchange rates for all major currencies in an hourly fashion, although this time size can be adjusted (e.g. 36hours, monthly, etc.). To make out general trends, we calculate the mean on a rolling basis for the last 50, 100 and 200 days. To increase confidence in the trader's decision, we also display Bollinger Bands for each data point. Additionally, we compare the forex buy and sell prices of 5 different major banks, to ensure the trader gets the best rate for his trade. Our web app automatically displays all five rates and the cheapest buy / highest sell to minimize inconveniences for our users. finally, we display real time forex news, including the corresponding sentiment, to allow for a well-founded, informed decision.

How we built it

For the web application side, we utilized Dash to simplify the amount of request that needed to be sent to web severs. The Forex news is delivered through an API (forexnews.api.com), as is the sentiment analysis. To visualize our Forex data from 2019-2023, we use the Plotly Figure, Bar and Candle functions. The data manipulation is done using pandas and sampling techniques.

Challenges we ran into

We faced several challenges. Firstly, we ran into some library versioning problems. Secondly, none of us have ever used the Dash library, so it wasn't easy getting something up and running in the beginning.

Accomplishments that we're proud of

We are proud that we were able to display the financial data alongside important trading technicals. We are also very happy that we managed to integrate news updates as it makes our app feel more realistic and useful.

What we learned

It is important not only to be able to code, but to know how to apply those technical skills to solve real business problems. Furthermore, we are very thankful to have a cooperative team that played to each other's strengths and communicated frequently. It is okay to fail. sometimes, code might not work despite multiple bug reviews. the important thing is to keep going and solve the problem/find a workaround)

What's next for PerspectiveFX

Optimisation. The application currently handles a massive number of data points on a single machine. There are many avenues for improvement. Additionally, integration of live trading information from brokerage APIs would make our application even more useful.

Built With

Share this project:

Updates