Financial data prediction method and device, storage medium and electronic device

Abstract

The invention discloses a financial data prediction method and device, a storage medium and an electronic device. The financial data prediction method comprises the steps that multiple pieces of financial data are acquired, internal association among the multiple pieces of financial data is learned from multiple aspects (inflection points and overall trends), multiple pieces of financial data association diagrams are constructed, one node in each piece of financial data association diagram corresponds to one piece of financial data, and each piece of financial data in the financial data association diagrams corresponds to one piece of financial data; the edge between the nodes represents the relevancy between the two pieces of data, and the larger the relevancy is, the thicker the edge is; in the designed prediction model, one or more graph convolutional neural networks are correspondingly configured for each graph, change trends of financial data are captured from different dimensions, and finally, through a multi-graph feature fusion layer and a full connection layer, multi-dimensional features are fused and prediction data of each piece of financial data in multiple pieces of financial data are output. The technical problem of low financial data prediction accuracy in the prior art is solved.

Description

technical field [0001] The present invention relates to the field of finance, and in particular, to a financial data prediction method, device, storage medium and electronic device. Background technique [0002] Stock market forecasting is a classic problem at the intersection of finance and computer science. The stock market forecasting model has two directions: fundamental analysis and technical analysis. Fundamental analysis evaluates the stock price according to the intrinsic value of the stock price, that is, the fair value. Most of the information is organized and analyzed from the perspectives of macroeconomics, industry analysis, and corporate conditions, while technical analysis is based on three major assumptions. First, it is assumed that market behavior includes all The second is to assume that there is an inertial trend, and the third is to think that history will repeat itself, and use logic, statistics and other methods to analyze market behavior as the resea...

AI Technical Summary

Problems solved by technology

[0006] Embodiments of the present invention provide a financial data forecasting method, device, storage medium, and electronic device to at least solve the technical problem of low financial data forecasting accuracy in the prior art

Images