Skip to main content

2016 | Buch

Technical Analysis for Algorithmic Pattern Recognition

insite
SUCHEN

Über dieses Buch

The main purpose of this book is to resolve deficiencies and limitations that currently exist when using Technical Analysis (TA). Particularly, TA is being used either by academics as an “economic test” of the weak-form Efficient Market Hypothesis (EMH) or by practitioners as a main or supplementary tool for deriving trading signals. This book approaches TA in a systematic way utilizing all the available estimation theory and tests. This is achieved through the developing of novel rule-based pattern recognizers, and the implementation of statistical tests for assessing the importance of realized returns. More emphasis is given to technical patterns where subjectivity in their identification process is apparent. Our proposed methodology is based on the algorithmic and thus unbiased pattern recognition. The unified methodological framework presented in this book can serve as a benchmark for both future academic studies that test the null hypothesis of the weak-form EMH and for practitioners that want to embed TA within their trading/investment decision making processes. ​

Inhaltsverzeichnis

Frontmatter
Chapter 1. Technical Analysis
Abstract
In this chapter the topic of technical analysis is discussed. Initially, technical analysis and its relation with the efficient market hypothesis are presented. Subsequently, a bundle of celebrated tools, that technicians implement in their trading activities, along with the corresponding, reported in the literature, empirical findings are presented. Particular emphasis is given on the bibliography on technical patterns, since this is the main area that this book examines. Afterwards, this chapter discusses on some controversial perceptions on the characterization and the implementation of technical analysis. At the end of the chapter, an outline of the book is given.
Prodromos E. Tsinaslanidis, Achilleas D. Zapranis
Chapter 2. Preprocessing Procedures
Abstract
The purpose of this chapter is to present two important preprocessing procedures than need to be carried before someone moves to the phase of recognizing technical patterns in financial price series. Initially, the importance of detecting errors in a dataset and various manners of replacing missing values are discussed.
This chapter further describes the algorithms for the identification of regional locals and perceptually important points. These algorithms are a common prerequisite step for all the proposed, patterns’ identification mechanisms presented in this book.
Prodromos E. Tsinaslanidis, Achilleas D. Zapranis
Chapter 3. Assessing the Predictive Performance of Technical Analysis
Abstract
This chapter presents some of the celebrated means by which the predictive performance of a technical trading system or a particular technical tool can be assessed. Although not all of these procedures are used in the subsequent chapters, we believe that they are important basic tools for anyone who wishes to assess the performance of such trading systems.
Prodromos E. Tsinaslanidis, Achilleas D. Zapranis
Chapter 4. Horizontal Patterns
Abstract
This chapter deals with horizontal technical patterns. After presenting existed identification techniques for these patterns, this chapter focuses on horizontal support and resistance levels. For these levels, our proposed identification mechanism is presented. The novelty of our methodology resides in the manner it encloses principles, found in well-known technical manuals, used for the identification of these levels via visual assessment. Their drawing derives from historical locals, rather than denoting support (resistance) levels from the lowest (highest) price levels of precedent constant time intervals. Finally this chapter assesses whether these levels are efficient trend-reversal predictors, and if they can generate systematically abnormal returns.
Prodromos E. Tsinaslanidis, Achilleas D. Zapranis
Chapter 5. Zigzag Patterns
Abstract
The purpose of this chapter is to present identification algorithms for a bundle of celebrated zigzag technical patterns and assess their performance. More precisely, this chapter presents identification mechanisms for the Head and Shoulders, Double Tops/Bottoms, Triple Tops/Bottoms, Flags, Pennants and Wedges technical patterns, and evaluates their performance on simulated and real datasets.
Prodromos E. Tsinaslanidis, Achilleas D. Zapranis
Chapter 6. Circular Patterns
Abstract
This chapter presents our proposed, novel, rule-based mechanism for identifying the rounding bottoms (also known as saucers) and tops patterns. These patterns are perhaps the most celebrated among circular technical pattern, and their identification is considered a difficult task to implement due to their circular characteristics. Their performance is further assessed in a manner similar with the one implemented in the previous chapter.
Prodromos E. Tsinaslanidis, Achilleas D. Zapranis
Chapter 7. Technical Indicators
Abstract
This chapter presents a bundle of well-known technical indicators along with the corresponding formulas for their calculation. More precisely, this chapter focuses on moving averages, moving averages crossovers, moving average convergence/divergence, relative strength index, Bollinger bands, momentum, price rate-of-change, highest high and lowest low indicators. This chapter completes the universe of technical tools that will be assessed collectively in the subsequent chapter.
Prodromos E. Tsinaslanidis, Achilleas D. Zapranis
Chapter 8. A Statistical Assessment
Abstract
This chapter examines the performance of TA, by implementing all technical patterns and indicators presented previous chapters, on 18 financial market indices around the world, for the requested period of 1998–2014. Parameters used for defining these trading rules are excerpted from the literature, and are the most commonly used. The methodology applied includes ordinary statistical tests and a bootstrap assessment.
Prodromos E. Tsinaslanidis, Achilleas D. Zapranis
Chapter 9. Dynamic Time Warping for Pattern Recognition
Abstract
This chapter presents a Dynamic Time Warping (DTW) algorithmic process to identify similar patterns on a price series. This methodology initially became popular in applications of voice recognition, and it is not considered to be included within the context of TA. In this chapter our analysis on technical pattern recognition processes is extended, by presenting an alternative methodology, which combines two main modifications of the DTW algorithmic process; the Derivative DTW and the subsequence DTW. We believe that this methodology captures the same conventions of technical patterns; that history is repeated, forming patterns which may vary in length. With this chapter we intend to inspire the reader to look for alternative quantitative techniques for recognizing similar patterns on financial price series, beyond those presented within the context of technical analysis.
Prodromos E. Tsinaslanidis, Achilleas D. Zapranis
Metadaten
Titel
Technical Analysis for Algorithmic Pattern Recognition
verfasst von
Prodromos E. Tsinaslanidis
Achilleas D. Zapranis
Copyright-Jahr
2016
Electronic ISBN
978-3-319-23636-0
Print ISBN
978-3-319-23635-3
DOI
https://doi.org/10.1007/978-3-319-23636-0