Mining Candlesticks Patterns on Stock Series: A Fuzzy Logic Approach

Authors:
Mario Linares Vásquez;Fabio Augusto González Osorio;Diego Fernando Hernández Losada
Affiliations:
Economic Optimization Research Group, National University of Colombia, Bogotá;Intelligent Systems Research Lab, National University of Colombia, Bogotá;Economic Optimization Research Group, National University of Colombia, Bogotá
Venue:
ADMA '09 Proceedings of the 5th International Conference on Advanced Data Mining and Applications
Year:
2009

Citing 2
Cited 0

Pattern Discovery of Fuzzy Time Series for Financial Prediction

IEEE Transactions on Knowledge and Data Engineering
The use of data mining and neural networks for forecasting stock market returns

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

Candlesticks is a technique used in financial time series in order to forecast future market performance. With candlesticks patterns, traders build active trading strategies in order to buy, sell or hold securities. The process is based on a preliminary stage which consists in identifying individual basic shapes on time series. Identifying candlesticks basic shapes is easy for a human, but recognizing complex patterns is hard because a lot of data is available. In this paper a data mining model for building active trading strategies (using candlesticks assumptions) is proposed looking for frequent itemsets on symbolic stocks series. Model validation is achieved with real data from New York Stock Exchange.