Self-adaptive similarities and differences moving average line stock trend prediction method

Abstract

The invention relates to the technical field of stock analysis, and discloses a self-adaptive similarities and differences moving average line stock trend prediction method, which comprises the following steps of S1, data division; s2, constructing a clustering standard; s3, clustering is carried out; s4, determining different parameters; training and learning the clustered oscillogram classification through a neural network, and determining different parameters for each category; s5, constructing a self-adaptive similarities and differences moving average line; establishing an MACD model foreach type of data, and combining all the MACD models to construct a final adaptive similarities and differences moving average line; and S6, stock prediction is carried out by using the adaptive similarities and differences moving average line, and according to the adaptive similarities and differences moving average line stock trend prediction method, different parameters are determined for different types of stocks, so that the constructed dynamic MACD index has relatively high applicability, and the prediction accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of stock analysis, in particular to an adaptive similarity-difference sliding average stock trend prediction method. Background technique [0002] Moving Average Convergence and Divergence (MACD, Moving Average Convergence and Divergence) is a technical analysis tool for judging buying and selling opportunities and tracking futures and stock market price trends. It is of great significance to stock price trend forecasting. How to enhance the adaptability of MACD and the accuracy of stock forecasting is the key to stock technical analysis. Among them, the effectiveness of stock data clustering and the accuracy of each type of data parameters are the key technical links in the construction of adaptive MACD, which still need to be further improved. [0003] The construction of existing MACD indicators mainly has the following shortcomings in terms of technology: [0004] (1) The applicability to different typ...

AI Technical Summary

Problems solved by technology

[0004] (1) The applicability to different types of stocks is not high

Figure 1 is the waveform diagram of the closing prices of two different stocks. The shapes and trends of the two waveform diagrams in Figure 1(a) and Figure 1(b) are completely different. If the MACD indicator with fixed parameters is used for prediction, it is bound to be different from the actual result. will produce a large error

[0005] (2) The applicability to different periods of the same class of stocks is not strong

figure 2 It is a waveform chart of the closing price of the same stock in different time periods. It can be seen from the figure that the stock price changes in the first period of the stock are relatively flat, but the fluctuations in the second half period are very violent, so it is inevitable to use the MACD indicator with fixed parameters to predict the development trend of the stock. Affects accuracy

Images