Improving trend reversal estimation in forex markets under a directional changes paradigm with classification algorithms

The majority of forecasting methods use a physical time scale for studying price fluctuations financial markets. Using scales can make companies oblivious to significant activities in the market as flow is discontinuous, which could translate missed profitable opportunities or risk exposure. Directional changes (DC) has gained attention recent years by translating series event-based series. Under this framework, trend reversals be predicted using length events. Having knowledge allows traders take an action before such happen and thus increase their profitability. In paper, we investigate how classification algorithms incorporated process predicting create DC-based trading strategies. effect proposed reversal estimation measured on 20 foreign exchange markets over 10-month period total 1000 data sets. We compare our results across 16 algorithms, both DC non-DC based, technical analysis buy-and-hold. Our findings show that introduction leads return higher profit statistically outperform all other