Inspiration
We love challenges, especially those in financial markets and quantitative strategies. Optiver's challenge was the perfect fit for us as it simulated real life's market dynamics: it allowed us to create innovative strategies to improve market's efficiency.
What it does
We created a trading algorithm that simulates the behavior of a market maker acting on a renewable energy market. It tries to improve its efficiency by first reducing the bid/ask gap, and then it tries to hedge using arbitrage of the market.
How we built it
We used Optiver's platform to model this algorithm, it fetches data from this simulated market and tries to exploit it by adjusting our positions. It looks for volume asymmetries and other's trader's positions to predict a sensible price point.
Challenges we ran into
Before coming up with the final algorithm we encountered many different dead ends. It was hard to let a cool idea go when it was not possible to make it work in practice.
Accomplishments that we're proud of
During the development process, we managed to implement nice adaptive functions. For example, we ideated a simple model that decides when to relax soft constraints and when it is possible to fastly return to a non-over-extended position on the market.
What we learned
We learned how simple trading algorithms work, and how big of an impact market makers have on the world.
What's next for GOAT_Trading_Algorithm
We would love to improve on this algorithm and try to deploy it in the real market.
Built With
- algorithmic-trading
- amazon-web-services
- python
Log in or sign up for Devpost to join the conversation.