Foreign exchange prediction and trading using Restricted Boltzmann Machines

Abstract

The exchange rate, the exchange ratio between currencies, can be regarded as the currency expression of the value of countries to each other. As an important link in international financial relations, the fluctuation of the exchange rate has a huge impact on the country's economy and the international balance of payments. Therefore, exchange rate predicting has always been a topic that has attracted attention in the technological explosion, and the importance of exchange rate prediction does not need to be questioned. The main goal of this project is to use restricted Boltzmann machines to forecast exchange rates. Exploring the internal pattern of the data through Restricted Boltzmann Machines can get the internal structure of the data. According to the application of Restricted Boltzmann Machines, it is used as feature extraction, to find the deep characteristics of the data. Then, the additional time series prediction model is used to forecast. To form comparison, the Deep Belief Network is added as a contrast.