Hybrid Forex prediction model using multiple regression, simulated annealing, reinforcement learning and technical analysis

Foreign exchange market refers to the in which currencies from around world are traded. It allows investors buy or sell a currency of their choice. Forex interests several categories stakeholders, such as companies that carry out international contracts, large institutional investors, via main banks, transactions on this for speculative purposes. One most important aspects is knowing when invest by buying, selling, and through recorded trend pair, but given characteristics namely its chaotic, noisy not stationary nature, prediction becomes big challenge traders it comes predicting accuracy. This paper aims predict right action be taken at certain moment development model combines multiple techniques regression, simulated annealing meta-heuristics, reinforcement learning technical indicators.