Research Repository

Developing trading strategies under the Directional Changes framework, with application in the FX Market

Bakhach, Amer (2018) Developing trading strategies under the Directional Changes framework, with application in the FX Market. PhD thesis, University of Essex.


Download (2MB) | Preview


Directional Changes (DC) is a framework for studying price movements. Many studies have reported that the DC framework is useful in analysing financial markets. Other studies have suggested that, theoretically, a trading strategy that exploits the full promise of the DC framework could be astonishingly profitable. However, such a strategy is yet to be discovered. In this thesis, we explore, and consequently provide proof of, the usefulness of the DC framework as the basis of a profitable trading strategy. Existing trading strategies can be categorised into two groups: the first comprising those that rely on forecasting models; the second comprising all other strategies. In line with existing research, this thesis develops two trading strategies: the first relies on forecasting Directional Changes in order to decide when to trade; whereas the second strategy, whilst based on the DC framework, uses no forecasting models at all. This thesis comprises three original research elements: 1. We formalize the problem of forecasting the change of a trend’s direction under the DC framework. We propose a solution for the defined forecasting problem. Our solution includes discovering a novel indicator, which is based on the DC framework. 2. We develop the first trading strategy that relies on the forecasting approach established above (Point 1) to decide when to trade. 3. We develop a second trading strategy which does not rely on any forecasting model. This is trading strategy employs a DC-based procedure to examine historical prices in order to discover profitable trading rules. We examine the performance of these two trading strategies in the foreign exchange market. The results indicate that both can be profitable and that both outperform other DC-based trading strategies. The results additionally suggest that none of these two trading strategies outperforms the other in terms of profitability and risk simultaneously.

Item Type: Thesis (PhD)
Uncontrolled Keywords: Foreign Exchange; Forecast; Algorithmic Trading; Directional Changes; Computational Finance
Subjects: H Social Sciences > HG Finance
Divisions: Faculty of Science and Health > Computer Science and Electronic Engineering, School of > Centre for Computational Finance and Economic Agents
Depositing User: Amer Bakhach
Date Deposited: 05 Dec 2018 12:01
Last Modified: 05 Dec 2018 12:01

Actions (login required)

View Item View Item