Bitcoin price prediction based on ARIMA-LSTM model

Abstract

As a decentralized digital currency, the price of Bitcoin is affected by multiple factors and has complex and non-linear characteristics. Traditional time series forecasting methods such as ARIMA models have limitations in dealing with these characteristics. In order to overcome these problems, a prediction algorithm based on the ARIMA-LSTM combined model is proposed. This algorithm captures the linear trend of Bitcoin through ARIMA model, and then models the nonlinear features and time dependence through LSTM model to improve the accuracy of prediction. Experimental results show that compared with a single model, the ARIMA-LSTM combination model has better prediction performance when dealing with highly volatile assets such as Bitcoin, which provides a good foundation for digital currency risk management and risk management in the financial market. It provides new ideas for investment decisions.