1. Intro
You want to think like a ‘data scientist’ who wants to manage his/her portfolio or hedge funds (the problem domain) in an investment firm. You will learn to design a trading strategy based on your predictions and backtesting. These predictions will come from Machine learning. Algo trading with ML will result in learning the following skills:
- Technology stack: Setting up a local environment and sourcing data
- Engineering financial features for predictive modeling (data manipulation and analysis)
- Linear regression model (It is ML machine learning!)
- Evaluating our ML model.
- Quality of ML predictions (performance analysis)
- Fine-tuning algorithms to improve results.
- Understand Algorithms’ results
- Backtesting a trading strategy
- Evaluating the performance of our trading strategy
- Evaluating the performance (report of this project)
- GO TO MODULE 1 (SETUP & PREPARATION)
2. Predictive features
This is the meat and potatoes of this project, to engineer financial features, from ‘technical indicators’ about future returns. In plain English it is data manipulation and analysis.
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nunc ullamcorper condimentum ultrices. Cras euismod ornare laoreet.
By using machine learning, I can predict and manage risk in trading portfolios by preemptively identifying potential losses and thus minimizing the impact of negative market movements.
3. Design
Build, tune, and test an ML model to predict asset returns for a given price horizon.ML can add value at multiple steps in the lifecycle of a trading strategy, and relies on key infrastructure and data resources
4. Design
Use cases of ML for trading
- Data mining for feature extraction and insights
- Supervised learning for alpha factor creation and aggregation
- Asset allocation
- Testing trade ideas
- Reinforcement learning
5. BUILD A TRADING STRATEGY
Develop a trading strategy by defining rules that translate model predictions into trades.
Predictive modeling
Sentiment analysis
Pattern recognition
Portfolio Optimization
Risk management
To use machine learning for algorithmic trading, you can:
1. Optimize trading strategies by using algorithms that interact with the market and adjust their strategy based on performance feedback
2. Predict future market movements using time series models and adjust trading strategies accordingly
3. Leverage NLP and deep learning to extract tradeable signals from market and alternative data.
6. Backtest
Step five resulted in building a trading strategy, now it’s time to back-test the strategy.
The incorporation of an investment idea into an algorithmic strategy requires extensive testing with a scientific approach that attempts to reject the idea based on its performance in alternative out-of-sample market scenarios. Testing may involve simulated data to capture scenarios deemed possible but not reflected in historical data.
7. EVALUATE
Finally, we are ready to evaluate the result using performance and risk measures.