The Equity Fund Risk Predictions Via Quantum-Classical Hybrid Neural Networks

Among the recent studies on the use of machine learning methods to study fund stocks, there are few studies on the use of quantum-classical hybrid models. In this paper, we investigate the application of a quantum-classical hybrid neural network algorithm in equity fund risk prediction. Specifically, we evaluate the performance of quantum-classical hybrid neural networks for stock fund risk prediction. In our experiments, we collected historical data from several stock fund portfolios. And then, we use the collected data to do the appropriate processing and encode the feature data onto quantum states. Subsequently, we utilized the proposed quantum-classical hybrid neural network to predict the risk indices of the funds. We measure the performance and accuracy of the model by calculating and counting the mean square error of the prediction results against the actual results of the test set, as predicted by the above algorithm. The experimental results show that the quantum-classical hybrid neural network proposed in this paper can achieve better risk prediction for stock funds, that is, the model can capture the nonlinearity between the input features of stock data and risk indices.