Journal of Economics and Management Volume 18, No. 1 March, 2022

Innovative Candlestick Technical Trading Strategies Using Genetic Algorithms

Ya-Chi Huang

Department of International Business, Lunghwa University of Science and Technology, Taiwan

Rueiher Tsaur

Department of Computer Information and Network Engineering, Lunghwa University of Science and Technology, Taiwan

Kuo-An Tseng

Department of Finance, Lunghwa University of Science and Technology, Taiwan

Abstract

The issue of whether the candlestick analysis method is profitable has been extensively discussed in academia in recent years, and the related studies have all been published in influential journals. Early studies regarding candlestick analysis methods focused on the six types of two-day solid candlestick patterns. Although the studies can be viewed as a milestone in the literature, the hit ratios for the proposed buy signals were only slightly higher than 50%. In this paper, we use genetic algorithms to evolve the candlestick technical trading strategies, which can effectively reduce the complexity of the information processing and enables us to consider factors that may affect the price trend more comprehensively. We incorporate the “rolling forward method” and split the data into a training period, a validation period, and a testing period to prevent the overfitting problem. To the best of our knowledge, we are the first to combine genetic algorithms with the candlestick analysis method to create innovative trading strategies.

Keywords：Candlestick Analysis Method, Genetic Algorithms, Artificial Intelligence,Technical Analysis.

JEL Classifications：G11, C63.

BACK