An Advanced CNN-LSTM Model for Cryptocurrency Forecasting

Abstract

Nowadays, cryptocurrencies are established and widely recognized as an alternative exchange currency method. They have infiltrated most financial transactions and as a result cryptocurrency trade is generally considered one of the most popular and promising types of profitable investments. Nevertheless, this constantly increasing financial market is characterized by significant volatility and strong price fluctuations over a short-time period therefore, the development of an accurate and reliable forecasting model is considered essential for portfolio management and optimization. In this research, we propose a multiple-input deep neural network model for the prediction of cryptocurrency price and movement. The proposed forecasting model utilizes as inputs different cryptocurrency data and handles them independently in order to exploit useful information from each cryptocurrency separately. An extensive empirical study was performed using three consecutive years of cryptocurrency data from three cryptocurrencies with the highest market capitalization i.e., Bitcoin (BTC), Etherium (ETH), and Ripple (XRP). The detailed experimental analysis revealed that the proposed model has the ability to efficiently exploit mixed cryptocurrency data, reduces overfitting and decreases the computational cost in comparison with traditional fully-connected deep neural networks.

Deep Learning for SARS COV-2 Genome Sequences

“…Thus, we use the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$5 \times 2$ \end{document} -fold cv paired \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}${t}$ \end{document} -test to perform a post-hoc analysis to determine the statistical significance of the differences in the means of the performance metric scores. Following [12] , [65] , we chose the AUC ROC as a specific measure to choose the model that would be more accurate on new test data. The test statistic \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$\tilde {t}$ \end{document} , for the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$5 \times 2$ \end{document} -fold cv paired \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}${t}$ \end{document} -test is calculated using the following equation [63] \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}\begin{equation*} \tilde {t} = \frac {p_{1}^{(1) }}{\sqrt {\frac {1}{5}\sum _{i = 1}^{5}s_{i}^{2}}}\tag{7}\end{equation*} \end{document} where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$p_{1}^{(1) }$ \end{document} is the difference in the AUC ROC scores of the CNN-Bi-LSTM vs CNN or CNN-LSTM models for the first fold of the first iteration, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$s_{i}^{2}$ \end{document} is the variance of the AUC ROC score differences of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}${i}$ \end{document} th iteration.…”

“…Thus, we use the 5 × 2-fold cv paired t-test to perform a post-hoc analysis to determine the statistical significance of the differences in the means of the performance metric scores. Following [12], [65], we chose the AUC ROC as a specific measure to choose the model that would be more accurate on new test data. The test statistict, for the 5×2-fold cv paired t-test is calculated using the following equation [63]…”

Prediction of Cryptocurrency Price Index Using Artificial Neural Networks: A Survey of the Literature

“…As shown in the previous sections, the purpose of this survey paper is to present and compare multiple research papers that employed multiple artificial neural network-based approaches to predict cryptocurrency prices. Noting the pros and cons of each of the methods presented (in terms of time elapsed, prediction accuracy, MAPE, and R-squared), it appears that in a holistic manner the methods used in Jaquart et al (2021), and the MICDL method in Livieris et al (2021) are of significant efficiency.…”

“…A similar framework is used in Livieris et al (2021), where the authors use a Convolutional Neural Networks (CNN) hybrid with LSTM to predict the prices of the three cryptocurrencies with highest market capitalization: Bitcoin (BTC), Ethereum (ETH), and Ripple (XRP). Their experiment implements three years (January 2017 to October 2020) of price data into a multiple-input deep neural network framework.…”

Is It Possible to Forecast the Price of Bitcoin?

“…Besides, Li et al [5] demonstrate that the Attentive LSTM network and an Embedding Network achieve superior state-of-the-art performance among all baselines for the Bitcoin price fluctuation prediction problem. Last but not least, Livieris et al [6] utilizes as inputs different cryptocurrency data and handles them independently to exploit helpful information from each cryptocurrency separately, which leads to better results than the traditional fully-connected deep neural networks.…”