Prediction of Bitcoin Exchange Rate to American Dollar Using Artificial Neural Network Methods

Published in 9th International Conference on Advanced Computer Science and Information Systems (ICACSIS), Jakarta, 2017

Abstract

Cryptocurrency trade is now a popular type of investment. Cryptocurrency market has been treated similar to foreign exchange and stock market. However, because of its volatility, there is a need for a prediction tool for investors to help them consider investment decisions for cryptocurrency trade. Nowadays, Artificial Neural Network (ANN) computing based tools are commonly used in stock and foreign exchange market predictions. There has been much research about ANN predictor on stocks and foreign exchange as case studies but none on cryptocurrency. Therefore, this research studied variety of ANN method to predict the market value of one of the most used cryptocurrency, Bitcoin. The ANN methods will be used to develop model to predict the close value of Bitcoin in the next day (next day prediction). This study compares four ANN methods, namely backpropagation neural network (BPNN), genetic algorithm neural network (GANN), genetic algorithm backpropagation neural network (GABPNN), and neuro-evolution of augmenting topologies (NEAT). The methods are evaluated based on accuracy and complexity. The result of the experiment showed that BPNN is the best method with MAPE 1.998 ± 0.038 % and training time 347 ± 63 seconds.

Recommended citation: Radityo, A., Munajat, Q., & Budi, I. (2017). Prediction of Bitcoin Exchange Rate to American Dollar Using Artificial Neural Network Methods. 9th International Conference on Advanced Computer Science and Information Systems (ICACSIS).

Download paper here