A fuzzy decision system for money investment in stock markets based on fuzzy candlesticks pattern recognition

摘要

This article proposes a novel fuzzy recommendation system for stock market investors. This intelligent decision tool uses fuzzy Japanese candlesticks and includes the effect of currency devaluation on the forecasting. To do so, first the next market session is obtained by a new designed fuzzy forecasting trading system. Then, it is compared to the one obtained by a non-parametric system based on the k-nearest neighbor technique. Finally, an amount of money to be invested is considered using a new capital management fuzzy strategy. The results have been compared to an analogous fuzzy trading system that has the same all-or-nothing investment strategy with risk control, but where this capitalization is not included. Both intelligent decision systems have been applied to two very different stock markets, the American Nasdaq100 and the Spanish lbex35 markets, using the Buy and Hold investment strategy as benchmark. Results prove that the proposed fuzzy system with capitalization is profitable and presents high stability, and could be a good support system for investors. (C) 2019 Elsevier Ltd. All rights reserved.