The FundBase challenge captured our attention, we always had curiosity on how the stock market and/or hedge founds worked. When we found out there was a chace to get our hands on a realistic problem we dived straight in.
What it does
Given a LARGE amount of raw data about hedge funds and a minimum fund size, minimum track record, maximum volatility, number of funds to invest in... it returns the optimal funds to be invested in and what percentage to invest in each of them.
How we built it
We split the work to be done:
- On one side, Isaac created the server, the web page and the link between the R code and the web page.
- On the other side, Adrià and Joan focused on getting the data from Amazon Dynamo DB, creating the mathematical model and applying it with RStudio.
Challenges we ran into
A lot. Starting from scratch in all of the techniques applied in this project meant a lot of challenges. Cleaning the raw data was tedious, setting up flask correctly (which worked after trying a second time something that didn't work before...) was time consuming.
Accomplishments that I'm proud of
First web page ever made, first time using python, flask, html, css, js. First time doing a large project in R.
What I learned
As always, first timers entail a lot of learning. I guess it can be summed up with:
What's next for StartHack Hedge Fund Advisor
The algorithm could be improved with more time. We could add better graphics to the web page for those who don't rely on numbers. Improve the data display since we didn't manage to do it in such little time.