Candlestick chart based trading system using ensemble learning for financial assets

The candlestick charts, which were developed in the 18th century and initially used Japanese rice market, are widely trading strategies all financial markets after 1991. Candlestick charts can interpret opening, high, low closing values of an asset a single visual. In addition to these advantages, large number chart patterns makes their practical use difficult. study for this purpose, software framework that uses predicts trend direction was created. consists four stages. first step, system recognizes candle is second stage, performance model measured by running training testing processes on data sets types labeled. machine learning phase, community methods such as xgboost used. last stage study, it seen with strategy based only recognizing pattern taking position proposed approach, higher profit obtained 11 world indices compared Buy&Hold strategy.