Inspiration

In the dynamic world of trading, the ability to make informed decisions quickly is paramount. We were inspired by the challenge of creating an algorithm that could not only interpret market data but also factor in real-world events, such as social media sentiment, to make trading decisions. The goal was to blend traditional trading indicators with modern data sources to gain a competitive edge.

What it does

Our trading algorithm, "Optiver Trading", is designed to:

Analyze market data in real-time, focusing on bid-ask spreads, volume differences, and price movements. Incorporate sentiment analysis from social media feeds, particularly tweets, to gauge market sentiment and predict potential price movements. Utilize traditional trading indicators like the Exponential Moving Average (EMA) and the Relative Strength Index (RSI) to identify market trends. Dynamically adjust bid and ask prices based on the algorithm's position, the EMA, and the impact of tweets. Ensure market liquidity and balance before making trading decisions. How we built it We started by connecting to the Exchange to fetch real-time market data. Using Python, we implemented various functions to calculate trading indicators like EMA and RSI. We also developed a mechanism to poll new social media feeds and assess their impact on stock prices. The core of our algorithm then uses these indicators and sentiment scores to decide on the trading direction and adjust bid/ask prices accordingly.

Challenges we ran into

Handling incomplete order books and ensuring that our algorithm can still make informed decisions. Integrating real-time sentiment analysis and determining its weight in our trading decisions. Avoiding self-trades and ensuring that our orders do not conflict with existing positions. Balancing between traditional trading indicators and the influence of social media sentiment. Accomplishments that we're proud of Successfully integrating sentiment analysis into a trading algorithm. Creating a dynamic system that adjusts its strategy based on multiple data sources. Ensuring that our algorithm remains robust even in the face of incomplete or conflicting data.

What we learned

The importance of blending traditional trading strategies with modern data sources. The challenges of real-time data processing and decision-making in trading. The intricacies of market dynamics and the factors that can influence stock prices.

What's next for Optiver Trading

Further refine the sentiment analysis component to consider more diverse data sources and improve accuracy. Implement machine learning models to predict price movements based on historical data and real-time indicators. Expand the algorithm to consider other trading indicators and strategies for a more holistic approach. Test and optimize the algorithm in various market conditions to ensure consistent performance.