Research On Early Warning System For Tilapia Trade

Based on the theories of industrial development, sustainable development, industrial safety, economic monitoring and early warning, also after consulted the researches of international trade development, trade safety of agricultural products, trade development of tilapia industry, and agricultural monitoring and early warning, this paper summarized the purpose and significance of early warning on trade of China’s tilapia industry, also strive to explore the impact of factors affecting the trade of China’s tilapia industry. Indicators of early warning system are the prerequisite and basis for building a prediction or evaluation model, which includes many aspects related to comprehensive competitiveness during the sustainable development of an industry. The first step is to select index, in accordance with the characteristics of China’s tilapia industry on trade, this paper established a indicators system by selected four quantitative index, included contribution rate of production, growth rate of trade volume, growth rate of price, international market share, and one qualitative index—olicy environment factor. According to different types of data, we selected principles of arithmetic mean, median and mode, ensure the calculate methods of warning level and warning limit of each index, also the calculate method of warning level of comprehensive early warning index. Based on Gray system theory, this paper built a GM(1,1) predictive model, combine with BP neural network predict method to predict the changes of all leading indicators. Then working out the warning level of warning situation indicators, and complete the establishment of early warning system on tilapia trade. The main use of Gray system is to study "a small sample with part information unknown", an uncertainty system with "poor information" by exploring, developing and searching all available information, then describe the operation behavior and change laws of early warning system correctly and monitor it reasonably. BP neural network calculate method is the way of output expectations based on the input-output mapping relation schema by learning and storing a large number of sample data. Analysis predicted value and evaluated weight of each warning situation indicators, then calculate the value and warning level of comprehensive early warning index, warning level of comprehensive early warning index will reflect the trade development trend of China’s tilapia industry in the future. Though a comprehensive analysis and compare the empirical results, we found:in recent years, the trade development of China’s tilapia industry is exceptional, industrial trade safety continued to be showed a warning status. This mainly reflected in the big fluctuate of trade volume and price of tilapia products, trade policy of main tilapia trade nations is becoming more and more inclined to protect their own industries, trade value to international market also has undergone major changes.Innovative ideas of this paper are shown as the following:(1) the thesis explore the various factors that affect trade of China’s tilapia products, first present to establish a early warning system on tilapia trade. It is the necessary tools and methods of the tilapia industry modern management to ensure the normal operation of tilapia trade.(2)The thesis first use Gray system and BP neural network theory and methods to establish a prediction model of leading indicators of early warning system, compare to normal regression analysis methods they would be more accurately to reflect the trade security situation of China’s tilapia products.