Abstract
Candlestick chart patterns are widely used in stock trading decisions. Patterns of candlestick series are found to provide hints for the price of the next one. This research proposes a technique to generate stock trading strategies employing features which are shown to effectively recognize patterns in candlestick charts. The features are combined into a tree-like trading strategy using the Chi-square Automatic Interaction Detector algorithm. The technique is evaluated using actual stocks from Stock Exchange of Thailand. The results show that the generated strategies are more profitable than other popular trading techniques, such as moving average convergence divergence, exponential moving average, relative strength index, stochastic oscillator and average directional index.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
