This book explains the broad topic of automated trading, starting with its mathematics and moving to its computation and execution. Readers will gain a unique insight into the mechanics and computational considerations taken in building a backtester, strategy optimizer, and fully functional trading platform.

Automated Trading with R provides automated traders with all the tools they need to trade algorithmically with their existing brokerage, from data management, to strategy optimization, to order execution, using free and publically available data. If your brokerage’s API is supported, the source code is plug-and-play.

The platform built in this book can serve as a complete replacement for commercially available platforms used by retail traders and small funds. Software components are strictly decoupled and easily scalable, providing opportunity to substitute any data source, trading algorithm, or brokerage. The book’s three objectives are:
To provide a flexible alternative to common strategy automation frameworks, like Tradestation, Metatrader, and CQG, to small funds and retail traders.
To offer an understanding the internal mechanisms of an automated trading system.
To standardize discussion and notation of real-world strategy optimization problems.
What you’ll learn

Programming an automated strategy in R gives the trader access to R and its package library for optimizing strategies, generating real-time trading decisions, and minimizing computation time.
How to best simulate strategy performance in their specific use case to derive accurate performance estimates.
Important machine-learning criteria for statistical validity in the context of time-series.
An understanding of critical real-world variables pertaining to portfolio management and performance assessment, including latency, drawdowns, varying trade size, portfolio growth, and penalization of unused capital.
Who This Book Is For

This book is for traders/practitioners at the retail or small fund level with at least an undergraduate background in finance or computer science. Graduate level finance or data science students.

Part 1: Problem Scope
Chapter 1: Fundamentals of Automated Trading
Equity Curve and Return Series
Characteristics of the Equity Curve
Characteristics of the Return Series
Risk-Return Metrics
Characteristics of Risk-Return Metrics
Sharpe Ratio
Maximum Drawdown Ratios
Partial Moment Ratios
Regression-Based Performance Metrics
Optimizing Performance Metrics
Part 2: Building the Platform
Chapter 2: Networking Part I
Yahoo! Finance API
Setting Up Directories
URL Query Building
Data Acquisition
Loading Data into Memory
Updating Data
YQL Web Service
URL and Query Building
Note on Quantmod
Background
Comparison
Organizing as Date-Uniform zoo Object
Note on zoo Objects
Chapter 3: Data Preparation
Handling NA Values
Note: NA vs. NaN in R
IPOs and Additions to S&P 500
Merging to the Uniform Date Template
Forward Replacement
Linearly Smoothed Replacement
Volume-Weighted Smoothed Replacement
Discussion of Replacement Methods
Real Time vs. Simulation
Influence on Volatility Metrics
Influence on Trading Decisions
Conclusion
Closing Price and Adjusted Close
Adjusting for Stock Splits
Adjusting for Cash Dividends
Efficient Updating and Adjusted Close
Implementing Adjustments
Test for and Correct Inactive Symbols
Computing the Return Matrix
Chapter 4: Indicators
Indicator Types
Overlays
Oscillators
Accumulators
Pattern/Binary/Ternary
Machine Learning/Nonvisual/Black Box
Example Indicators
Simple Moving Average
Moving Average Convergence Divergence Oscillator (MACD)
Bollinger Bands
Custom Indicator Using Correlation and Slope
Indicators Utilizing Multiple Data Sets
Conclusion
Chapter 5: Rule Sets
Our Process Flow as Nested Functions
Terminology
Example Rule Sets
Overlays
Oscillators
Accumulators
Filters, Triggers, and Quantifications of Favor
Chapter 6: High-Performance Computing
Hardware Overview
Processing
Multicore Processing
Hyperthreading
Memory
The Disk
Random Access Memory (RAM)
Processor Cache
Swap Space
Software Overview
Compiled vs. Interpreted
Scripting Languages
Speed vs. Safety
Takeaways
for Loops vs. apply Functions
for Loops and Memory Allocation
apply-Style Functions
Use Binaries Creatively
Note on Measuring Compute Time
Multicore Computing in R
Embarrassingly Parallel Processes
doMC and doParallel
The foreach Package
The foreach Package in Practice
Integer Mapping
Computing the Return Matrix with foreach
Computing Indicators with foreach
Chapter 7: Simulation and Backtesting
Example Strategies
Our Simulation Workflow
Listing 7-1: Pseudocode
Listing 7-1: Explanation of Inputs and User Guide
Discussion
Implementing Example Strategies
Summary Statistics and Performance Metrics
Conclusion
Chapter 8: Optimization
Cross Validation in Time Series
Numerical vs. Analytical Optimization
Numerical Optimization Overview
Parameter Transform for Unbounded Search Algorithms
Declaring an Evaluator
Listing 8-1: Pseudocode
Listing 8-1: Explanation of Inputs and User Guide
Exhaustive Search Optimization
Pattern Search Optimization
Generalized Pattern Search Optimization
Nelder-Mead Optimization
Nelder-Mead with Random Initialization
Projecting Trading Performance
Conclusion
Chapter 9: Networking Part II
Market Overview: Brokerage APIs
Secure Connections
Establishing SSL Connections
Proprietary SSL Connections
HTTP/HTTPS
OAuth
Feasibility Analysis for Trading APIs
Feasibility of Custom R Packages
HTTPS + OAuth Through Existing R Packages
FIX Engines
Exporting Directions to a Supported Language
Planning and Executing Trades
The PLAN Job
The TRADE Job
Common Data Formats
Manipulating XML
Generating XML Documents
Manipulating JSON Data
The Financial Information eXchange Protocol
The FIX eXtensible Markup Language
OAuth in R
Conclusion
Part 3: Production Trading
Chapter 10: Organizing and Automating Scripts
Organizing Scripts into Jobs
Calling Jobs with the Source Function
Calling Jobs via Sourcing
Task Scheduling in Windows
Running R from the Command Line in Windows
Setting Up and Managing the Task Scheduler
Task Scheduling in UNIX
Conclusion
Chapter 11: Looking Forward
Language Considerations
Python
C/C++
Hardware Description Languages
Retail Brokerages and Right to Refuse
Right to Refuse in the Swiss Currency Crisis
Connection Latency
Ethernet vs. WiFi
Proximity to Exchanges
Prime Brokerages
Digesting News and Fundamentals
Conclusion
Appendix A: Source Code
Platform/config.R
Platform/load
Platform/load.R
Platform/update.R
Platform/functions/yahoo.R
Platform/load/initial.R
Platform/load/loadToMemory.R
Platform/load/updateStocks.R
Platform/load/dateUnif.R
Platform/load/spClean.R
Platform/load/adjustClose.R
Platform/load/return.R
Platform/load/fillInactive.R
Platform/compute
Platform/compute/MCinit.R
Platform/compute/functions.R
Platform/plan
Platform/plan.R
Platform/plan/decisionGen.R
Platform/trade
Platform/trade.R
Platform/model
Platform/model.R
Platform/model/optimize.R
Platform/model/evaluateFunc.R
Platform/model/optimizeFunc.R
Appendix B: Scoping in Multicore R
Scoping Rules in R
Using Lexical Scoping
Takeaways
The UNIX fork System Call
The fork Call and Memory Management
Scoping Implications for R
Instance Replication in Windows
Instance Replication and Memory Management
Scoping Implications for R
Index