Generalized stock price prediction method based on multitask asymmetric proximity support vector machine

Abstract

The invention provides a generalized stock price prediction method based on a multitask asymmetric proximity support vector machine, and solves the problems of poor prediction precision and robustnesscaused by relatively high limitation due to neglect of cross-correlation information among stock indexes and distribution characteristics of a data set in an existing stock market prediction method.The prediction method mainly comprises the following steps: obtaining and preprocessing a plurality of stock trading day quotation data to obtain a data set; constructing an asymmetric square epsiloninsensitive loss function, and adjusting hyper-parameters to enable the hyper-parameters to better adapt to the distribution characteristics of the data set; establishing a generalized mathematical optimization model for stock price prediction based on the multitask learning asymmetric proximity support vector machine according to the multitask learning hypothesis; converting an original planningproblem of a generalized stock price prediction model based on the multitask learning asymmetric proximity support vector machine into a dual planning problem according to a KKT condition; and solvingthe dual planning to obtain a decision function of the stock price prediction model.

Description

technical field [0001] The application relates to a data processing method suitable for forecasting commercial and financial trend information, and specifically relates to a stock price forecasting method. Background technique [0002] The stable development of the stock market is the focus of attention of the national government and investors, and the analysis of the running trend of the stock market index and the fluctuation of stock prices has become a widely discussed topic in the academic and financial circles. So far, scholars at home and abroad have done a lot of research on stock market forecasting. However, due to the characteristics of the stock market itself, such as complex nonlinearity, high dimensionality, noise sensitivity, and instability, it is particularly important to conduct reasonable and accurate volatility prediction analysis. difficulty. [0003] The existing stock market prediction methods are mainly divided into three categories: market technical a...

AI Technical Summary

Problems solved by technology

[0006] The purpose of this application is to solve the problem of poor prediction accuracy and robustness due to the large limitations of existing stock market forecasting methods due to the neglect of cross-correlation information between stock indexes and the distribution characteristics of data sets

Images