Financial time series data are characterized by non-linearity, non-stationarity, and stochastic complexity, so predicting such data presents a significant challenge. This paper proposes a novel hybrid model for financial forecasting based on CEEMDAN-SE and ARIMA- CNN-LSTM. With the help of the CEEMDAN-SE method, the original data are decomposed into several IMFs and reconstructed via sample entropy into a lower-complexity stationary high-frequency component and a low-frequency component. The high-frequency component is predicted by the ARIMA statistical forecasting model, while the low-frequency component is predicted by a neural network model combining CNN and LSTM. Compared to some classical prediction models, our algorithm exhibits superior performance in terms of three evaluation indexes, namely, RMSE, MAE, and MAPE, effectively enhancing model accuracy while reducing computational overhead.
19 October 2022 in Acta Mathematicae Applicatae Sinica, English Series
In today’s data-driven world, efficient and secure cross-institution information-sharing is an urgent challenge. Traditional information-sharing methods based on access controlling often suffer from issues such as privacy breaches and high communication complexity. To address this issue, this paper proposes a cross-institution information-sharing solution based on a consortium blockchain, in which it combines on-chain transaction consensus with off-chain institution storage, thereby facilitating collaboration among nodes from different institutions on the blockchain. To enhance the efficiency and security of transactions on the blockchain, we also introduce a dynamic and adaptive Practical Byzantine Fault Tolerance (DA-PBFT) consensus protocol, which permits nodes to dynamically join and exit the blockchain network, consequently improving network scalability. Through a reputation mechanism, we swiftly identify and remove faulty and malicious nodes, enhancing the trustworthiness of nodes in the information-sharing network based on consortium blockchain, thereby improving consensus efficiency. We have also employed encryption techniques to enhance the privacy and integrity of data during the process of cross-institution information sharing. A comprehensive analysis of the communication complexity in the information-sharing network confirms the effectiveness and security of our proposed solution. We offer a unique solution to improve the efficiency and security of cross-institution information-sharing while ensuring data integrity and privacy. By addressing the challenges of privacy breaches and high communication complexity in information sharing, we establish a foundation for secure cross-institution data exchange.
In this paper, the insider trading model of Xiao and Zhou (Acta Mathematicae Applicatae, 2021) is further studied, in which market makers receive partial information about a static risky asset and an insider stops trading at a random time. With the help of dynamic programming principle, we obtain a unique linear Bayesian equilibrium consisting of insider's trading intensity and market liquidity parameter, instead of none Bayesian equilibrium as before. It shows that (i) as time goes by, both trading intensity and market depth increase exponentially, while residual information decreases exponentially; (ii) with average trading time increasing, trading intensity decrease, but both residual information and insider's expected profit increase, while market depth is a unimodal function with a unique minimum with respect to average trading time; (iii) the less information observed by market makers, the weaker trading intensity and market depth are, but the more both expect profit and residual information are, which is in accord with our economic intuition.
A stochastic discrete fractional Cournot duopoly game model with a unique interior Nash equilibrium is developed in this study. Some sufficient criteria of the Lyapunov stability in probability for the proposed model at the interior Nash equilibrium are derived using the Lyapunov theory. The proposed model’s finite time stability in probability is then investigated using a nonlinear feedback control approach at the interior Nash equilibrium. The stochastic Bellman theory is also used to explore the locally optimum control problem. Furthermore, bifurcation diagrams, time series, and the 0-1 test are used to investigate the chaotic dynamics of this model. Finally, numerical examples are given to illustrate the obtained results.
16 October 2024 in Applied Mathematics and Computation