Research On Risk Early Warning Of "Going Out" Of China’s Agricultural Enterprises Based On Bayesian Network

Agricultural "going out" is a major measure for China to promote the development of modern agriculture,coordinate international and domestic resources,ensure the supply of food security and practice the community of human destiny,and has received positive responses from all parties since it was proposed.Nowadays,China’s agricultural "going out" tends to be specialized,centralized and diversified,and the number of agricultural enterprises participating in them is increasing,but it is facing a more complex international environment,and the corresponding risk management problems are becoming increasingly prominent.This thesis focuses on the risks encountered by agricultural enterprises "going out",systematically analyzes the relationship between risk variables and quantitatively evaluates and provides early warning,applies machine learning algorithms to agricultural risk management,and uses literature research,expert knowledge,survey interviews,questionnaire surveys and empirical analysis to conduct qualitative and quantitative research on the risks of "going out" of agricultural enterprises,and establishes a risk index system and a Bayesian network risk early warning simulation model.The risk value and early warning light number are used to intuitively reflect the risk state of the risk system,and then further verify and apply the model through the simulation analysis of the research cases,and then put forward the risk prevention suggestions for agricultural enterprises to "go global",helping agricultural enterprises to use limited resources to achieve the best risk management effect.The main contents of the study are as follows:(1)Discussed the motivation and risk causes of China’s agricultural enterprises "going out",based on the needs of strategic development,agricultural enterprises actively participate in "going out",but due to the lack of experience accumulation and risk awareness,their "going out" process is not smooth,coupled with the weak ability of the agricultural industry to cope with risks,there are strict safety access standards and trade protection policies,so that agricultural enterprises "going out" risk management faces more challenges.(2)From eight risk dimensions,namely product risk,operation and management risk,financial risk,natural environment risk,political risk,legal risk,market risk and cultural risk,the influencing factors of agricultural enterprises "going out" are specifically analyzed from two internal and external perspectives.(3)Create a risk questionnaire,form a risk index system according to the reliability and validity test results,standardize the sample data through the risk level matrix,apply expert knowledge and experience to build a more realistic topology structure to describe the influence relationship between risk factors,and then use Netica software for machine learning to generate Bayesian network model,set risk lights and early warning intervals,and build a complete simulation model of agricultural enterprise "going out" risk warning.According to the above research,it is concluded that:(1)the overall risk of the initial state of the "going out" risk system of agricultural enterprises shows a red light,and the risk value of the risk dimension Chinese risk and the risk management risk is higher,which is displayed as a red light,while the early warning light number of exchange rate changes,lack of international experience,trade protection problems,price fluctuations and unwritten rules in the risk factor is black,and the risk level is extremely high and requires special attention.(2)The effectiveness of the model is verified by applying the positive reasoning prediction of the research case to the model,and then the reverse reasoning of the overall risk and the risk of each dimension is used to diagnose the degree of influence of the relevant risk factors,and the risk importance ranking is carried out,such as the risk value of the default risk in the legal risk is higher than the risk value of the legal complexity and the risk value of the incoterms standard issue of the contract.(3)In view of the main risk problems of the simulation model early warning,six risk prevention suggestions are put forward,and agricultural enterprises should conduct in-depth and comprehensive research on the "going out" project and prepare detailed feasibility analysis reports;adopt diversified cooperation methods to expand and extend the industrial chain;form teams with domestic enterprises to "go out" and establish local enterprise alliances;pay attention to the information and consulting services released by the government,and provide professional protection through third-party institutions;pay attention to corporate social responsibility and maintain good relations with local governments,residents and partners Rational allocation of enterprise management resources,enhance the strength of enterprise international operation.In summary,agricultural enterprises can apply the risk early warning model to monitor the risk status of "going out" projects,and lay out relevant risk management measures in advance to help enterprises develop benignly.