Research On The Profitability Of Artificial Intelligence Listed Companies Based On Panel Data

In 2017,China included "artificial intelligence" into the government report for the first time.In 2018,the number of Chinese AI enterprises was second only to that of the United States.The transformation of old and new economic drivers promoted the acceleration of the growth of the emerging industry of artificial intelligence.However,in the capital market,most of the rise has been accompanied by amplified information,and in recent years,financial panic and the negative blow of the Sino-US trade war have required listed AI companies to prepare for a rainy day.Through analysis,this paper believes that paying close attention to the profitability of listed AI companies is an important means for operators to grasp the lifeblood of companies and investors to grasp the direction of investment.In this paper,through literature review,found that most of the research of the listed company profit ability of the literature is through cross section data to carry on the empirical analysis,and artificial intelligence optimal development of listed companies in recent years,so the selection panel data to explore the profitability of listed companies of artificial intelligence can be combined with the characteristics of time series data and cross section data,observation samples more accurately reflect differences in the group and the difference between groups.With the in-depth development of the relevant model of panel data,it is the general trend to study the profitability of listed AI companies based on panel data.Most of the research about the profitability of listed companies is through cross section data to carry on the empirical analysis,and artificial intelligence optimal development of listed companies in recent years,choose the panel data of recent years to explore the profitability of listed companies of artificial intelligence can be combined with the characteristics of time series data and cross section data,observation samples more accurately reflect differences in the group and the difference between groups.With the in-depth development of relevant models of panel data,it is an irresistible trend to study the profitability of listed AI companies based on panel data.This paper first addresses how to comprehensively and comprehensively measure the profitability of listed AI companies.In order to avoid one-sidedness of single index and subjectivity of multi-index artificial weight,this paper adopts the multi-index factor analysis method of panel data to solve the problem.The common factors are solved by dimensionality reduction processing with two data transformation methods.Through the comparison of results,the method of using cross-section multi-index factor analysis for each year's data is more suitable for the solution of this problem.Finally,the method is used to integrate the information of eight indicators in four aspects to obtain the comprehensive index of the profitability of listed AI companies.It is not enough to understand the profitability of listed AI companies,but also how to improve the profitability of listed companies.In order to have a more comprehensive understanding of the specific effects of various influencing factors on the level of profitability at different loci,this paper constructs a panel quantile regression model for random effects,and uses panel Bayesian LASSO quantile regression method for parameter estimation,and identifies redundant variables to make the more accurate valuation model more robust.Finally,through empirical analysis,this paper selects two redundant variables,namely current ratio and current asset turnover rate,and analyzes the influence direction and trend of various influencing factors on the profitability of listed AI companiesFinally,combined with the development status of listed AI companies and empirical research results,this paper puts forward relevant Suggestions to provide theoretical guidance for the long-term development of listed AI companies in China.