Short-Term Forecasting in Bitcoin Time Series Using LSTM and GRU RNNs

doi:10.5753/kdmile.2020.11964

Paper

Short-Term Forecasting in Bitcoin Time Series Using LSTM and GRU RNNs

Published Oct 20, 2020 · Marcelo De Caux, F. Bernardini, José Viterbo

Anais do Symposium on Knowledge Discovery, Mining and Learning (KDMiLe 2020)

UNKNOWN SJR score

7

Citations

1

Influential Citations

PDF

Semantic Scholar

Abstract

In recent years, bitcoin has become a very attractive investment in financial industry, which is not controlled by governments, but is based on trust between transfers under the technology of block chain. Hence, forecasting future bitcoin cryptocurrency values is a problem that has attracted the attention of many researchers in the field, while proving to be a very challenging problem. This work presents an experimental analysis using LSTM and GRUs for forecasting bitcoin values in a minute-granulated time for the entire next day. To this end we also present our methodology for conducting the experiments. The final goal is to create the core of a financial prediction tool around the RNNs. In our experiments, we achieved interesting results such as a SMAPE of 0.0002, a RMSE of US$ 3.844 and a rRMSE of 0.0028 in a day where bitcoin rates vary from US$ 13.2K and US$ 14.6K, surpassing the results of SMAPE found in the literature and proposed limit of SMAPE smaller than 0.007 for forecasts.

Study Snapshot

Key takeawayUsing LSTM and GRU RNNs, we can effectively forecast bitcoin values in a minute-granulated time for the entire next day, surpassing previous results and literature limits.

PopulationOlder adults (50-71 years)

Sample size24

MethodsObservational

OutcomesBody Mass Index projections

ResultsSocial networks mitigate obesity in older groups.

Sign up to use Study Snapshot

Consensus is limited without an account. Create an account or sign in to get more searches and use the Study Snapshot.

Create an account

Paper

Short-Term Forecasting in Bitcoin Time Series Using LSTM and GRU RNNs

Sign up to use Study Snapshot

References

Bitcoin Financial Forecasting

Linear Regression is the most accurate algorithm for predicting the value of Bitcoin in USD compared to Support Vector Regression and Neural Network Regression.

Forecasting Bitcoin closing price series using linear regression and neural networks models

Our models, using linear regression and neural networks, can accurately forecast Bitcoin, Litecoin, and Ethereum daily closing prices, outperforming benchmarks and improving the state-of-the-art.

A Comparison between ARIMA, LSTM, and GRU for Time Series Forecasting

ARIMA model outperforms deep learning-based regression models in time series forecasting, with GRU outperforming LSTM.

Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019

Deep Learning models significantly outperform traditional Machine Learning models in financial time series forecasting, with potential for future advancements.

Crypto-Currency price prediction using Decision Tree and Regression techniques

Decision tree and regression models both accurately predict Bitcoin prices, with the decision tree model showing better performance than the regression model.

Citations

On the benefits of automated tuning of hyper-parameters: an experiment related to temperature prediction on UAV computers

Automated tuning of hyper-parameters in neural networks can improve temperature prediction accuracy by up to 36.02% compared to exhaustive search, achieving satisfactory results with a reduced dataset.

Deep Learning Model Analysis and Web-Based Implementation of Cryptocurrency Prediction

This study developed a web-based application using hybrid GRU LSTM to predict cryptocurrency prices for the next few days, providing valuable information for cryptocurrency investment.

Deep Learning Based Resource Availability Prediction for Local Mobile Crowd Computing

Convolutional LSTM models effectively predict user availability in mobile crowd computing, minimizing job offloading and job loss due to user mobility.