BASIC REQUIREMENTS FOR TRADING ON CHART PATTERN RECOGNITION

Millions of investors base their investment decisions on computer models that generate trading signals on either fundamental or technical data. The smartest and highest-paid programmers try to come up with trading models with only one goal in mind: to make money every year. Few trading models reach this goal, and those that do are not available to the public. The reason is simple: If something really works, people better keep quiet, because otherwise either too much money gets involved or system followers eventually destroy the performance. The huge advantage of using computerized strategies is that computers are emotionless, fast, accurate, and disciplined.

In this computer-driven era, trading signals have to be generated by computers in order to be accepted. People think that the bigger and faster the computer, the better the trading signals. This is wrong. There is no guarantee that computer-generated trading signals will produce profitable trading signals. The saying still applies: Garbage in, garbage out. The question is not how good historical test results look, but how reliable the trading signals that produce the historical test results will be in future real-time trading.

One of the stable elements in analyzing and working with price data is investor behavior. People around the world and their investment decisions drive the prices of financial instruments on the exchanges up and down. Peaks and valleys are formed ...

Get Trading with Charts for Absolute Return now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.