Skip to content
All library documents

A Multi-Scale Residual CNN and LSTM for Bitcoin Price Forecasting

Article arXiv papers · Author: Qiutong Guo et al.
Read the original

Summary

The paper proposes MRC-LSTM, a model combining a multi-scale residual convolutional module with a Long Short-Term Memory network to predict Bitcoin’s daily USD closing price. One-dimensional convolutions are used to detect and combine features across different time scales in multivariate series, while the LSTM learns longer-term dependencies. Inputs include Bitcoin trading information as well as external factors such as macroeconomic variables and investor attention.

Experiments compare the model with other network structures and report that it performs significantly better for the Bitcoin prediction task. Additional experiments on Ethereum and Litecoin are presented as support for its use in short-term multivariate cryptocurrency forecasting. The supplied description does not give the data period, evaluation measures, detailed baselines, or out-of-sample trading results, so it does not show whether improved price prediction translates into profitable strategies or persists in other settings.

Key ideas

  • MRC-LSTM combines a multi-scale residual convolutional network with an LSTM for Bitcoin closing-price prediction.
  • The convolutional component detects and fuses features across time scales in multivariate data.
  • The LSTM component learns longer-term dependencies in the series.
  • The study includes trading data alongside macroeconomic variables and investor-attention measures.
  • Reported comparisons favor the proposed model, with further cryptocurrency experiments but no trading-performance evidence in the supplied description.

Tags

CryptoMachine learningStatisticsVolatility
Full text
# MRC-LSTM: A Hybrid Approach of Multi-scale Residual CNN and LSTM to Predict Bitcoin Price


# MRC-LSTM: A Hybrid Approach of Multi-scale Residual CNN and LSTM to Predict Bitcoin Price









Bitcoin, one of the major cryptocurrencies, presents great opportunities and challenges with its tremendous potential returns accompanying high risks. The high volatility of Bitcoin and the complex factors affecting them make the study of effective price forecasting methods of great practical importance to financial investors and researchers worldwide. In this paper, we propose a novel approach called MRC-LSTM, which combines a Multi-scale Residual Convolutional neural network (MRC) and a Long Short-Term Memory (LSTM) to implement Bitcoin closing price prediction. Specifically, the Multi-scale residual module is based on one-dimensional convolution, which is not only capable of adaptive detecting features of different time scales in multivariate time series, but also enables the fusion of these features. LSTM has the ability to learn long-term dependencies in series, which is widely used in financial time series forecasting. By mixing these two methods, the model is able to obtain highly expressive features and efficiently learn trends and interactions of multivariate time series. In the study, the impact of external factors such as macroeconomic variables and investor attention on the Bitcoin price is considered in addition to the trading information of the Bitcoin market. We performed experiments to predict the daily closing price of Bitcoin (USD), and the experimental results show that MRC-LSTM significantly outperforms a variety of other network structures. Furthermore, we conduct additional experiments on two other cryptocurrencies, Ethereum and Litecoin, to further confirm the effectiveness of the MRC-LSTM in short-term forecasting for multivariate time series of cryptocurrencies.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.