Correlation Analysis and Prediction of Stock Based on VMD-LASSO Model
QIN XIWEN, XU DINGXIN, Jiajing Guo · ECONOMIC COMPUTATION AND ECONOMIC CYBERNETICS STUDIES AND RESEARCH · 2021
For nonlinear non-stationary sequences, variational mode decomposition (VMD) is a novel, efficient, adaptive, quasi-orthogonal, completely non-recursive data decomposition method, which still has a solid theoretical basis.It iteratively searches for the optimal solution of the variational model to determine the frequency center and bandwidth of each component, so that the frequency domain segmentation of the signal and the effective separation of components can be adaptively realized.At the same time, the Lasso method is an effective method for performing variable screening.Therefore, this paper proposes a least absolute shrinkage and selection operator (LASSO) regression method based on the effective variable selection of components derived from VMD decomposition.The VMD-LASSO model is established for stock data.It is found that there is a strong interaction between the two stocks, and the influence of each component is one-toone.VMD-LASSO model is used to predict stock series, and the results are compared with those of three traditional methods.The results show that the proposed model has higher prediction accuracy.