Study On Early Warning Of Information Disclosure Violations Of The Listed Companies Of National Equities Exchange And Quotations In China Base On Genetic Neural Network

National Equities Exchange and Quotations(NEEQ)is an important part of the multi-level capital market in China.The listed companies of NEEQ should disclosure information in a truthful,accurate,complete and prompt manner.Once the information disclosure violations occurred,investors would suffer losses and the market would also face a crisis of confidence.Therefore,it is necessary to effectively warn the information disclosure violations of the listed companies.This paper selects the listed companies of NEEQ as research objects,which are punished by the regulators for information disclosure violations.Based on a comprehensive analysis of the current situation of violations,this paper starts with a probe into the motivations of violations empirically.By genetic neural network method,this paper establishes early warning models based on the indicators of company finance,market trading,governance and motivation,and conducts an early warning analysis of the listed companies of NEEQ.The main work and conclusions of this paper are as follows:Through empirical analysis and tests,it is found that shareholders' tunneling level,market manipulation level and earning management level are significantly positive correlated with the information disclosure violations of the listed companies of NEEQ.Shareholders' tunneling,market manipulation and earning management are the main motivations of the information disclosure violations.Based on the indicators of company finance,market trading,governance and motivation,early warning models for information disclosure violations and violation severity are established by genetic neural network method.By small sample tests,the accuracy rate of the early warning model for information disclosure violations reaches 83.64%,and the accuracy rate of the early warning model for violation severity reaches 73.24%.Moreover,it is found that the more indictors of company finance,market trading,governance and motivation the early warning models use,the higher accuracy rate they reach.The models are applied to make early warning empirical analysis on the listed companies of NEEQ in 2016.The warning effect of the models is good.By overall analysis,it is found that the number of the listed companies in the "violation" warning state is much larger than the number of those actually punished.By industry analysis,it is found that the number of the listed companies with information disclosure violations in manufacturing,information transmission,software,and information technology service industries is largest.The listed companies in mining,agriculture,forestry,animal husbandry,fishery,culture,sports,entertainment,finance and construction industries are more prone to information disclosure violations.The information disclosure violation severity of the listed companies in mining,agriculture,forestry,animal husbandry,fishery,wholesale and retail trades,leasing and business service industries is higher.Finally,from the perspective of strengthening the external regulatory mechanism and internal management mechanism,this paper provides policy recommendations to prevent the information disclosure violations of the listed companies of NEEQ.