Research on the valuation of City Enterprises Based on ARIMA series prediction

Abstract

This paper studies the valuation of listed companies on the science and technology innovation board. By establishing the corresponding mathematical model, comparing the stock market valuation level of China and the United States, through multiple data screening and analysis, using factor analysis of variance, time series model and RBF neural network analysis, the paper establishes the valuation and quantitative model of China’s stock market. AR. Regression analysis can accurately measure the correlation between each factor and improve the accuracy of the prediction method. Neural network has very strong nonlinear fitting ability, can map any complex nonlinear relationship, and the learning rules are simple, which is convenient for calculation and simulation. In order to quantitatively analyze the relationship between each factor and the valuation index, the significance test value of the Sino US market can be obtained by using the method of one-way ANOVA, and the ranking of the importance of each factor to the valuation index can be obtained through the analysis. The fundamental index and liquidity index of stock market have certain stability. In order to forecast and analyze the fundamental index and liquidity index of Chinese market and American market, the time series model is used to establish the prediction model for the change of index data.