| Due to the rapid development of blockchain technology,the types of digital cryptocurrencies are increasing day by day.At the same time,because the exchanges are very scattered,the trading risks of digital cryptocurrencies represented by Bitcoin have expanded.Predicting the price risk of digital cryptocurrencies can help market participants to conduct risk prevention.At the same time,digital cryptocurrency is an important means to realize the flow of digital assets,and the development of digital economy requires the support of more digital cryptocurrencies.The digital cryptocurrency transaction itself has the characteristics of high frequency,market fragmentation,high volatility,and freedom from time and space constraints.In traditional forecasting methods,these characteristics of digital cryptocurrency transactions cannot be effectively preserved,so the forecasting accuracy is low and the effect is poor.The long short-term memory network(LSTM)in deep learning can effectively solve nonlinear and long-term memory problem in digital cryptocurrency price risk prediction and improve the prediction effect.At the same time,the joint regression analysis of Value at Risk(Va R)and Expected Shortfall(ES)can effectively overcome the non-elicitability problem of ES,to better measure the risks and volatility of the financial market.By embedding LSTM into the joint regression combined forecasting framework of Va R and ES,a Va R and ES joint regression combined forecasting model based on LSTM is constructed to measure the risk of the Bitcoin trading market,that is,the LSTM-J-C model.The empirical results show that the newly proposed LSTM-J-C model can effectively improve the forecasting accuracy of Va R and ES in the Bitcoin market compared with the historical simulation,the GARCH model and the joint regression combined forecasting model that combines HS and GARCH.The research results can provide theoretical support and practical guidance for digital cryptocurrency market investors,policy makers and regulatory agencies to measure digital cryptocurrency market risks. |