IMG_3400.jpeg

# Algorithmic Trading ## Strategies and Their Automation ### Dr. Ernest P. Chan ### Chapter 1 ### Introduction #### What is Algorithmic Trading? Algorithmic trading is the execution of buy or sell orders generated by computer algorithms with little or no human intervention. The algorithms could be very simple, for example, "buy 100 shares of IBM at the close". Or they can be very complex, such as those used by Renaissance Technologies LLC, a quantitative hedge fund managing tens of billions of dollars. Algorithmic trading is also known as: automated trading, black-box trading, or simply algo-trading. #### Why Algorithmic Trading? * **Reduce transaction costs**: An algorithm can be designed to break up a large order into smaller pieces to minimize impact on the market. * **Obtain better prices**: By strategically timing the execution of orders, an algorithm can often obtain better prices than a human trader can. * **Reduce human errors**: Once an algorithm is programmed correctly, it will execute trades exactly as specified, without the possibility of human error, such as mis-keying an order. * **Backtesting**: Before an algorithm is deployed with real money, its performance can be rigorously tested on historical data. This is known as backtesting, and it can give the trader confidence that the algorithm will perform as expected in live trading. * **Trading opportunities**: Algorithms can be designed to identify and exploit trading opportunities that a human trader might miss. These may include arbitrage opportunities, statistical anomalies, or news events. * **Diversification**: Algorithmic trading can be used to diversify a portfolio by trading in multiple markets and asset classes simultaneously. * **Speed**: Algorithms can react to market events much faster than a human trader can. This is especially important in fast-moving markets. * **Discipline**: Algorithms can be programmed to follow a specific trading plan, without being influenced by emotions or biases. This can help traders to avoid making impulsive decisions. * **Improved Execution**: Algorithmic trading can improve the execution of trades by using sophisticated order routing and optimization techniques. #### Drawbacks of Algorithmic Trading * **Model Risk**: Backtesting results may not be indicative of future performance if the market changes. * **Data Mining Bias**: It is possible to find patterns in historical data that are purely random and have no predictive power. * **Overfitting**: An algorithm that is too closely fitted to historical data may not perform well in live trading. * **Technical Glitches**: Computer systems can fail, and algorithms can have bugs. * **Monitoring**: Algorithms need to be monitored to ensure that they are functioning correctly. #### Popularity In 2009, Tabb Group estimated algorithmic trading to account for 60% of US equity trading. #### Regulatory Scrutiny The rise of algorithmic trading has led to increased regulatory scrutiny, particularly after the "Flash Crash" of May 6, 2010, when the Dow Jones Industrial Average plunged nearly 1,000 points in a matter of minutes. #### Types of Algorithmic Trading Strategies * Trend Following. * Mean Reversion. * Arbitrage. * Statistical Arbitrage or "Stat Arb". * Market Making. * High-Frequency Trading ("HFT"). * Smart Order Routing. #### Trend Following Trend following strategies seek to capture profits from assets that are trending in a particular direction. These can be implemented based on: * Moving Averages * Breakouts * Channels Trend following is usually applied to liquid and volatile instruments, such as: * Stocks * Futures * Currencies * Bonds #### Mean Reversion A mean-reversion strategy assumes that prices tend to revert to their average value over time. These can be implemented based on: * Pairs Trading * Bollinger Bands * Stochastic Oscillators Mean reversion is usually applied to less liquid and less volatile instruments, such as: * Stocks * Bonds #### Arbitrage Arbitrage strategies seek to profit from price discrepancies between different markets or assets. These are based on the idea that the same asset should have the same price in all markets. Arbitrage can be implemented based on: * Triangular Arbitrage * Covered Interest Rate Parity * Index Arbitrage Arbitrage is usually applied to very liquid instruments, such as: * Currencies * Stocks * Futures #### Statistical Arbitrage Statistical arbitrage is a more sophisticated version of arbitrage that uses statistical models to identify and exploit price discrepancies. It's also known as "Stat Arb". Statistical arbitrage can be implemented based on: * Pairs Trading * Index Arbitrage * Factor Models Statistical arbitrage is usually applied to liquid instruments, such as: * Stocks * Futures * Currencies #### Market Making Market making strategies seek to profit from the bid-ask spread. Market makers provide liquidity to the market by standing ready to buy or sell an asset at any given time. Market making can be implemented based on: * Order Book Analysis * Inventory Management * Quote Management Market making is usually applied to liquid instruments, such as: * Stocks * Futures * Currencies #### High-Frequency Trading High-frequency trading (HFT) is a type of algorithmic trading that is characterized by high speed, high turnover, and high order-to-trade ratios. HFT firms use sophisticated algorithms and powerful computers to execute a large number of orders in a very short period of time. High-frequency trading can be implemented based on: * Order Book Analysis * News Analysis * Event Detection High-frequency trading is usually applied to very liquid instruments, such as: * Stocks * Futures * Currencies #### Smart Order Routing Smart order routing (SOR) is a type of algorithmic trading that seeks to find the best available price for an order by routing it to multiple exchanges or market centers. Smart order routing can be implemented based on: * Price Discovery * Order Routing * Execution Optimization Smart order routing is usually applied to liquid instruments, such as: * Stocks * Futures * Currencies #### Summary of Strategies | Strategy | Objective | Approach | Instruments | | :-------------------- | :-------------------------------------------- | :------------------------------------ | :------------------------- | | Trend Following | Capture profits from trending assets | Moving Averages, Breakouts, Channels | Stocks, Futures, Currencies | | Mean Reversion | Profit from price reverting to average value | Pairs Trading, Bollinger Bands | Stocks, Bonds | | Arbitrage | Exploit price discrepancies | Triangular Arbitrage, Covered Interest | Currencies, Stocks | | Statistical Arbitrage | Exploit statistical price discrepancies | Pairs Trading, Factor Models | Stocks, Futures, Currencies | | Market Making | Profit from bid-ask spread | Order Book Analysis, Quote Management | Stocks, Futures, Currencies | | High-Frequency | Exploit short-term market inefficiencies | Order book Analysis, Event Detection | Stocks, Futures, Currencies | | Smart Order Routing | Find best price by routing orders | Price Discovery, Execution Optimizat. | Stocks, Futures, Currencies | #### The Book The objective of this book is to provide a comprehensive guide to algorithmic trading strategies that can be realistically implemented, i.e. not requiring million-dollar infrastructure or exclusive access trading venues. This book will focus on the strategy and coding (in Python). Chapter X provides the reader with software and data vendors needed for this purpose.

Document Details

Related

Full Transcript