Quantum-Classical Hybrid Model Improves Financial Volatility Forecasting

Researchers propose a hybrid framework combining Long Short-Term Memory (LSTM) networks with Quantum Circuit Born Machines (QCBM) for financial volatility forecasting. The LSTM extracts temporal features, while the QCBM learns complex market distributions as a generative prior. Tested on 5-minute high-frequency data from SSE Composite and CSI 300 indices, the model outperforms classical LSTM baselines in MSE, RMSE, and QLIKE metrics. A stochastic 'Drop-Prior' mechanism is introduced to enhance robustness. The study highlights quantum computing's potential in high-dimensional sampling problems for finance.