Design and implementation of a quantitative investment system based on deep learning and data mining

Bitcoin, a representative virtual currency, has both investment risks and opportunities due to its price uncertainty and instability. Combining bitcoin with traditional financial products, such as gold, is a common hedging strategy for investors. Meanwhile, it is crucial for traders to accurately predict the price of financial products and make reasonable decisions, which can effectively help them avoid risks and maximize returns. The traditional financial measurement method is unable to capture the nonlinear of financial markets, multi-scale features, complex data, statistical machine learning needs a lot of expert knowledge, the artificial intervention, we choose the suitable for time series data in depth study of LSTM price prediction, and to prevent noise interference of financial data, we propose a VMD-LSTM prediction model. That is, THE IMF extracted from VMD decomposition and reflecting the internal law of the market is directly used as the feature for LSTM prediction.