Research On Pre-Warning Model Of Sustainable Development For China's Agricultural Listed Companies

As a respresentative in the advanced agricultrual production force in China, agricultural listed companies undertake the important task of promoting the economic development of agriculture. The achievement of strnegthening the agriculture as the foundation proposed by the state is largely depedent on the sustainable development and growth of agricultural listed companies. Therefore, whether agricultural listed companies can achieve sustainable development,which is directly related to the sustainable development of agriculture.However, in recent years, the sustainable development of agricultural listed companies is downward.The profitability, solvency,ability to grow, and the company's environmental and social performance, are in the weak state, while the environmental and social sustainable development are particularly prominent. In this context, it is necessary to commence the study of agricultural listed companies'sustainable development,and to enhance the sustainable development awareness of managent team and promote the companies'sustainability.Taking 51 agricultural listed companies as the sample,based on the basic economic theories,adopting the quantitative analysis as the main body and accopanied by the necessary finacial analysis, case studies and statistical analysis,this thesis has analysed the current status of agricultural listed companies and the influencing factors from the both macro and micro levels.The significant different indicators among three states (crisis, gray and good) are disclosd,and these indicators consist of the basic indicators of early warning model. This thesis has also discussed the influencing factors of early warning model's accuracy and the improvement direction, and how to commence sustainable development management combing the model, to enhance the sustainablity of agricultural listed companies. Specifically-this thesis has done researeh into the following issues:the first one is to construct a theoretical framework for corporate sustainability by defining the concept of sustainable development of enterprises and analysing of the stakeholder theory. The second one is objective description of the economic, social and environmental sustainability performance on the bisis of financial reports of agricultural listed companies(including Social Responsibility Report), and the analysis of influencing factors on agricultural listed companies. The third one is the designing an early warning indicator system, on the basis of the corporate sustainability analysis framework and influencing factors, by using mathematical statistics and factor analysis methods. The fourth one is the buliding the sustainable development eraly warning models of T-2 and T-3year, using the multivariate discriminant analysis and artificial neural network model approach, according to the eraly warning indcator system,and the testing of models'precision. The fifth one is of discussing the sustainable development management combing the model, as Luoniushan a case.The main conclusions can be drawn as follows:(1) Through the theoretical study and the disclosure of listed company report, we can find that the understanding of corporate sustainable development still remains in the narrow dimension.Corporate sustainable development means how to plan and manage corporate scientifically. That is to say, the corporate not only pursue the short-term operation results and market expansion, but also seek long-term sustainable profit growth, and enhance its competitive advantage and market value, in order to keep the business sustainable,while companies lack sufficient knowledge of sense of commitment to corporate citizenship.(2) According to the analysis of the sample companies'sustainable development, agricultural listed companies'sustainability is poor in China. The profitability, operating capacity and growth are not strong, showing an unstable trend, while the company's environmental and social issues are particularly highlighted in recent years.The factors influencing the sustainable development of listed companies include technology innovation, corporate governance, profitability and blind main business diversification, firm size, environmental and social sense of responsibility and so on. Meanwhile, the macroeconomic environment and economic policy, agricultural industry characteristics also have an impact on sustainable development of agricultural listed companies.(3) Through the mathematical statistical analysis,we can find that 14 indicators show significant differences among three kinds of agricultural listed companies both for T-2 year model and T-3 year models, and these indicators consist of the core pre-warning indicators.The number of T-3 early warning indicators is less than T-2 years, indicating that the significant indicators are less because of relatively longer time interval under different kinds of status of agricultrual listed companies, and this is in line with expectation.(4) According to the comparison analysis between multivariate discriminant analysis and BP neural network prediction model, the precision of early warning remains the same. While the success rate of early-warning model with T-2 year data is higher than the model with the data of T-3 year, about 3 and 7 percentage points are improved respectively. In addition, the success rate of the early-warning model is affected by the classification standard of corporate sustainability, sample amount and finiacial data, so the success rate can be improved from these pespectives. Corporate sustainbility management, combined with the early warning model of agricultrual listed company, which shows a good effect. The key innovations embody in:(1) combing the company sustainable development and the thougut of early warning model buliding, monitoring and pre-warning the sustainable development of agricultural listed companies, and carrying on sustainability management with the model,which is a new attempt. (2)based on the company sustainablity analysis framework to discuss the designing of early warning incator system, buliding the early warning model, doing an emprical test and a specific case applicition, which is full of new ideas.(3)designing the early warning indicator system by means of mathematical statistics and multivariate statistical methods, buliding T-2 and T-3 year sustainable development pre-warning models by by means of multivariate discriminant analysis and BP neural network model methods, which shows good effect.The shortcoming lies in three aspects:(1)the science and feasibility of assumptions on the status of agricultrual listed companies'sustainbility are to be further tested and studied.(2)the comprehension and effectiveness need to go through a long and a lot of practice tests before being fully recognized.(3) small sample size affects the representative of the early warning indicators, while the sample size is too small to cause difficulties in matching sample selection, and implementation of new accounting standards has an impact on the comparability of financial data, reducing the early warning accuracy of the model.