Trading Strategies using Python

This course examines different approaches to building trading strategies across all asset classes. Using Python you will learn how to interact with market data to perform data analysis and find trading signals. We will look at advanced strategies to maximize trade performance and examine the statistics around testing and evaluating trade performance.

The first day of the programme begins with an overview of trading strategies and outlines various approaches used to look for trading opportunities. Through workshops we will implement data analytics using Python standard libraries including Pandas and Scipy.

On day two we investigate how to generate trading signals and analyze some of the dangers of overfitting the data. A workshop will focus on how to implement these trading signals in practice using real data. The afternoon examines methods for evaluating the performance of the trading strategies alongside methods of execution. We finish by exploring more advanced trading strategies and Machine Learning techniques in trading.

• Traders
• Portfolio managers
• Fund managers
• Structurers
• Quantitative analysts
• Technologists and data scientists
• Risk managers

• Gain familiarity with the various forms of trading strategies
• Understand how to interact with standard Python libraries to perform data analysis
• Learn how to generate trading signals from data
• Perform back-testing and simulations and appreciate the risk of overfitting data
• Learn how to measure performance of strategies and transaction cost analysis
• Look into how machine learning techniques can be used to enhance trading signals
• Implement a trading strategy in Python using signals and optimize performance

A basic understanding of capital markets and securities trading. Some exposure to Python programming would be beneficial.

Introduction to Trading Strategies

• Fund investment
  • Active vs. passive strategies
  • Opportunity funds
• Statistical arbitrage
• Fundamental analysis
• Technical analysis
• Algorithmic trading
• High frequency trading
• Standard strategies
  • Trend following, Momentum
  • Mean reversion, moving averages, breakouts
  • Carry trading

Workshop: Python coding framework

• Installation, IDEs
• A rapid introduction to Python coding
• Built-in libraries: datetime, dateutil
• Python numerical libraries
  • Pandas
  • NumPy
  • SciPy
  • TensorFlow

Data Analysis

• Forms of data
  • Market data
  • Trade data
  • Order flow
  • Economic data and news
  • Alternative data
• Statistical approaches to data analysis

Workshop: Data and databases in Python

• Importing data
• Data formatting

Trading Signals

• What is a signal?
• Backtesting
• Dangers of overfitting data
• In-sample vs. out of sample
• Simulation and Monte Carlo
• A/B testing
• Walk-ahead testing
• Constructing stronger signals

Workshop: Generating trading signals in practice

• Data mining
• Hypothesis testing

Advanced Trading Strategies

• Execution strategies
  • Algorithmic execution
  • Venue choice
• Volatility trading
• Correlation and pairs trading

Trade Performance Metrics

• Trading against a benchmark
• Sharp ratio and maximum drawdown
• Best execution
• Transaction cost analysis

Machine Learning

• Machine learning methods for trading
• TensorFlow

Workshop: Trading against a signal

• Paper trade simulation