Skip to main content
Open Access Publications from the University of California


UCLA Electronic Theses and Dissertations bannerUCLA

Bitcoin Price Forecast Using LSTM and GRU Recurrent networks, and Hidden Markov Model


Bitcoin, the first decentralized cryptocurrency, has become popular not only because a growing size of merchants accepts it in transactions, but also because people buy it as an investment. This study focuses on the Bitcoin price forecast using Hidden Markov Model and two machine learning methods, LSTM and GRU Recurrent networks. Evaluated by MAPE and RMSE, the results indicate that the Hidden Markov Model with the Gaussian Mixture Models has the best performance among all methods. The GRU model outperforms the LSTM model, though sometimes it might have a more extreme result. When the price remains constant or changes steadily, the predictions are more precise than the fluctuation period.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View