Algorithmic Trading

Task 1 - Unsupervised Learning Trading Strategy

• Unsupervised Learning Trading Strategy, utilizing S&P 500 stocks data to master features, indicators, and portfolio optimization.

Features and technical indicators used for each stock.

• Garman-Klass Volatility
• RSI
• Bollinger Bands
• ATR
• MACD
• Dollar Volume

For each month select assets based on the cluster and form a portfolio based on Efficient Frontier max sharpe ratio optimization

• First we will filter only stocks corresponding to the cluster we choose based on our hypothesis.
• Momentum is persistent and my idea would be that stocks clustered around RSI 70 centroid should continue to outperform in the following month - thus I would select stocks corresponding to cluster 3.
• Then maximise the sharpe ratio with Efficient Frontier Optimizer.
• Apply diversification according to weight bound constraints.

Task 2 - Twitter Sentiment Trading Strategy

• Leveraged the power of social media with the Twitter Sentiment Investing Strategy, ranking NASDAQ stocks based on engagement and evaluating performance against the QQQ return.

Task 3 - Intraday GARCH Trading Strategy

• The Intraday Strategy will introduce you to the GARCH model, combining it with technical indicators to capture both daily and intraday signals for potential lucrative positions.