Stock trend prediction method and device, equipment and storage medium

Abstract

The invention discloses a stock trend prediction method and device, equipment and a storage medium. The method comprises the following steps: acquiring historical transaction data of a target stock; extracting feature parameters based on the historical transaction data; and inputting the characteristic parameters into a pre-trained hidden Markov model to enable the hidden Markov model to output the trend of the target stock in the target trading day. According to the technical scheme provided by the embodiment of the invention, the method improves the accuracy of stock trend prediction.

Description

Technical field [0001] The present application generally relates to the field of computer technology, specifically involving a stock trend prediction method, device, equipment and storage medium. Background [0002] Since the stock price is volatile, stock investors need to predict the next trend of the stock price before making an investment, and make corresponding trading decisions, so the prediction of stock trend is very important for a stock investor. [0003] At present, the general prediction of stock trend through the double moving average method, the specific process is: the establishment of m-day moving average, n-day moving average, and find the intersection of the two lines, if m >n, the n-day moving average crossing the m-day moving average moment is the golden cross moment, the time of the lower crossing of the m-day moving average is the dead fork moment, the long operation is carried out at the golden cross moment, and the short operation is carried out at the...

AI Technical Summary

Problems solved by technology

[0005] a) The double moving average method is difficult to accurately determine the time period of the two moving averages:

[0006] If the time period of the two moving averages is too close, the two moving averages will be intertwined and produce more intersections, which will cause a large number of trading signals to be generated by this method, and wrong trading signals are prone to appear;

[0007] If the time periods of the two moving averages differ greatly, it will be difficult to grasp the real-time information in the market, and it is easy to miss the best trading time;

[0008] b) For the case of large stock fluctuations in the short term, this method is prone to errors. Since the method is based on the average value within a certain period as the main timing index, if the fluctuations between the data in the short term are too large, it is easy to affect the entire stock market. The position of the moving average, thus generating false trading signals

Images