Quantitative Technical Analysis: An Integrated Approach to Trading System Development and Trade Management

by Howard B. Bandy

Quick Summary

A rigorous, code-oriented guide to developing and validating quantitative trading systems, covering data analysis, pattern recognition, non-stationary distributions, signal-to-noise ratios, position sizing, and systematic trade management. Emphasizes simulation-based validation and realistic performance expectations.

Detailed Summary

Howard B. Bandy presents a technically demanding, code-intensive guide to quantitative trading system development, published by Blue Owl Press in 2015. The book is explicitly educational in nature and includes downloadable program code for all examples, written for clarity rather than computational efficiency.

The Introduction establishes the goal: developing a systematic, evidence-based approach to trading that integrates system development with trade management. Bandy identifies major changes in the trading landscape that necessitate updated approaches and lays out the fundamental premises of technical analysis upon which the book's methodology rests.

The Data chapter addresses the critical but often overlooked foundation of any quantitative system. Bandy distinguishes between primary data (price, volume) and auxiliary data (economic indicators, sentiment measures, etc.), and emphasizes that traders should "assume nothing about distribution." The chapter examines data distributions in time sequence, lists of trades, distributions of trades, patterns, non-stationarity (the fact that market behavior changes over time), synchronization issues between different data series, signal versus noise decomposition, and the critical principle that data series are not interchangeable -- a system developed on one security cannot be assumed to work on another.

The Position and Trading System Development chapters form the core of the book. The position management framework addresses how to determine position size based on account equity, risk tolerance, and the statistical properties of the trading system. Trading system development follows a rigorous process: hypothesis formulation, indicator selection and parameter optimization, in-sample testing, out-of-sample validation, walk-forward analysis, and Monte Carlo simulation to establish confidence intervals around performance expectations.

Bandy places extraordinary emphasis on avoiding overfitting -- the tendency to develop systems that perform brilliantly on historical data but fail in live trading. He advocates for the minimum number of parameters consistent with capturing the intended market behavior, extensive out-of-sample testing, and realistic assumptions about execution costs including commissions, slippage, and delay between signal generation and order execution.

The trade management section covers entry timing, initial stop placement, trailing stop methodologies, profit target optimization, position scaling (pyramiding), and the integration of multiple systems within a portfolio context. Each topic is presented with complete program code and simulation results, allowing readers to replicate and extend the analysis on their own computers with their own data.

The book assumes significant programming competency and mathematical sophistication. It is not a book of ready-made trading systems but rather a toolkit and methodology for developing and validating systems rigorously. Bandy's academic background (Ph.D.) is evident in the disciplined approach to hypothesis testing and statistical validation that pervades every chapter.