Multivariate cryptocurrency prediction: comparative analysis of three recurrent neural networks approaches

Abstract

As a new type of currency introduced in the new millennium, cryptocurrency has established its ecosystems and attracts many people to use and invest in it. However, cryptocurrencies are highly dynamic and volatile, making it challenging to predict their future values. In this research, we use a multivariate prediction approach and three different recurrent neural networks (RNNs), namely the long short-term memory (LSTM), the bidirectional LSTM (Bi-LSTM), and the gated recurrent unit (GRU). We also propose simple three layers deep networks architecture for the regression task in this study. From the experimental results on five major cryptocurrencies, i.e., Bitcoin (BTC), Ethereum (ETH), Cardano (ADA), Tether (USDT), and Binance Coin (BNB), we find that both Bi-LSTM and GRU have similar performance results in terms of accuracy. However, in terms of the execution time, both LSTM and GRU have similar results, where GRU is slightly better and has lower variation results on average.

Forecasting cryptocurrency's buy signal with a bagged tree learning approach to enhance purchase decisions

“…Next, we collected cryptocurrency price data and applied it to develop technical indicators using a Python model to process the data collection. To evaluate market movements and develop trading signals, we use TA-LiB as a technical analysis library supporting statistical tools (Hansen et al, 2022 ). Then, we trained our model using the library to extract the technical signal.…”

Committer Assessment Practice in Blockchain Project: A Systematic Literature Review

“…In contrast, off-chain governance is a set of rules not encoded within the system, but that can be accomplished via community consensus (Reijers et al, 2018). Even if Blockchain projects do away with the requirement for potentially dishonest third parties (Hansun et al, 2022), developers, investors, and users are still faced with a governance crisis in the off-chain (De Filippi & Loveluck, 2016) where conflicting interests pose a risk to the Blockchain project as a whole (Bosu et al, 2019). Such a problem reveals the limitations of excessive dependence on technical means (on-chain) to handle social coordination and economic exchange difficulties in software projects (Reijers et al, 2018).…”

“…Blockchain technology demonstrates the essential characteristics of digital information systems (IS) artefacts established for cryptocurrency (Hansun et al, 2022). Since an anonymous individual or group of developers presented Bitcoin in 2008, there has been a rising interest in Blockchain projects (Bosu et al, 2019), with the peerto-peer networks of Bitcoin and Ethereum emerging as the two most well-known Blockchain initiatives (Biais et al, 2019).…”

“…The initial Blockchain project established a global decentralised currency and payment system that may be acceptable globally without the need for a central authority (Hansun et al, 2022). This radical financial innovation, foundationally based on the concept of opensource software (OSS), triggered the emergence of Decentralized Autonomous Organization (DAO) as a business model (Mehar et al, 2019).…”

A Deep Learning CNN based Approach to the Problem of Crypto-Currency Prediction

“…There was also a multivariate prediction on cryptocurrency done by Seng Hansun and others [12] who used multivariate prediction approach and used three different RNNs-LSTM, Bi-directional LSTMs and GRU on 5 major cryptocurrencies-Bitcoin, Etherum, Cardano, Tether, and Binance. Their research resulted in the conclusion that Bi-LSTM and GRU resulted in similar accuracy but with regards to execution time, GRU and LSTM depicted similar results.…”