A Financial Decision Supporting System Based on Fuzzy Candlestick Patterns

Abstract

A financial decision supporting system based on the fuzzy candlestick pattern is proposed and developed. We model Japanese candlestick patterns by using fuzzy linguistic variables. Japanese candlestick theory is an empirical model of investment decision. The theory assumes that the trend of financial time series could be predicted by identifying specific candlestick patterns in the candlestick charts. In our system, the investment expertise is represented in fuzzy candlestick patterns and can be stored in a pattern base. The investors can make their investment decisions based on the identified candlestick patterns. A knowledge based pattern recognition method for candlestick pattern is implemented in the system, and the investor can edit, validate, and share the imprecise and vague candlestick patterns through the graphical interface of the proposed system.