Stock price prediction method and device based on wavelet transform and long-term and short-term memory neural network fusion

A long short-term memory, price prediction technology, applied in the field of data information, can solve the problems of Fourier transform difficulty, time-frequency localization, etc., to solve the gradient disappearance and gradient explosion, reduce the prediction error, and solve the time-frequency localization. the contradictory effect of

Inactive Publication Date: 2019-12-24
SHANDONG UNIV OF FINANCE & ECONOMICS
View PDF0 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention effectively solves the contradiction of time-frequency localization and solves the difficult problem of Fourier transform by introducing variable scale factor and translation factor

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Stock price prediction method and device based on wavelet transform and long-term and short-term memory neural network fusion
  • Stock price prediction method and device based on wavelet transform and long-term and short-term memory neural network fusion
  • Stock price prediction method and device based on wavelet transform and long-term and short-term memory neural network fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] The present invention provides a stock price prediction method based on wavelet transform and long short-term memory neural network fusion, such as Figures 1 to 7 As shown, the methods include:

[0023] S1, perform wavelet decomposition on the stock sequence based on discrete wavelet transform;

[0024] Wavelet transform analysis is a sequence analysis method with time-frequency localization properties. The key is to decompose the sequence by wavelet. ψ(t) is a square integrable function, that is, ψ(t)∈L 2 , whose Fourier transform is like The following conditions:

[0025]

[0026] Then ψ(t) is called the base wavelet. After the basic wavelet undergoes stretching and translation operations, a new wavelet sequence is obtained, which is called a sub-wavelet:

[0027]

[0028] ψ a,b (t) is a wavelet basis function dependent on parameters a and b. Among them, a is the scaling factor or scale factor, and changing the size of a means doing scaling transform...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a stock price prediction method and device based on wavelet transform and long-term and short-term memory neural network fusion, and effectively solves the contradiction of time-frequency localization by introducing a variable scale factor and a translation factor compared with Fourier transform. Wavelet transformation is used for analyzing local information of frequency intime and space; multi-scale and multi-aspect refinement is carried out on the time sequence through stretching and translation operation, the purpose of frequency subdivision can be finally achieved,the requirement of time sequence analysis can be automatically met, the method can be focused on any aspect of the time sequence, and the problem of Fourier transform difficulty is solved. The LSTM has better performance in a longer sequence by increasing the connection of cell states, and effectively solves the problems of gradient disappearance and gradient explosion of the RNN along with the increase of the number of hidden layers. Wavelet transformation and a long-short-term neural network are fused, prediction errors are reduced through modeling of frequency information, and prediction precision is improved.

Description

technical field [0001] The present invention relates to the field of data information technology, in particular to a stock price prediction method and device based on wavelet transform and long-short-term memory neural network fusion. Background technique [0002] At present, the securities market is becoming more and more perfect, and more and more investors enter the stock market, trying to capture the potential model of the market. Affected by company decisions, government policies, cross-market breaking news and other factors, the stock market has strong volatility and instability, which also increases the difficulty of predicting future price trends. Stock price forecasting can understand and grasp the operating law of the stock market and the transmission mechanism of monetary policy. From a practical point of view, stock price forecasting can effectively select and implement monetary policy in the case of violent stock market fluctuations, which helps to reduce and e...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & AuthorityApplications(China)
IPC IPC(8): G06Q40/04G06Q10/04G06N3/04
CPCG06N3/049G06Q10/04G06Q40/04
Inventor刘慧刘鑫郭强张彩明
OwnerSHANDONG UNIV OF FINANCE & ECONOMICS