Early Warning System For SiChuan Industrial Economy

This article started from the economy early warning theory. It introduced the theoretic system of economy early warning, including setting up early warning index system, selecting early warning methods, determining warning degrees, and analyzing and removing potential risks. With early warning methods as the main clue, the article also introduced the development and current status of early warning in both domestic and international scope.With reference to the responsibilities and working flows of Sichuan Economy Committee, this article provided analysis on the indispensability of setting up the Sichuan Province industrial economy early warning system. It also analyzed the feasibility on the basis of human resource, material resources and fund resources of Sichuan Economy Committee. In the systems analysis, the article analyzed the operation data of Sichuan Economy Committee and generated the system UC matrix. It developed the subsystems and made definitions on their functions according to the matrix.As to the systematic structure, according to the characteristics of analysis oriented , the article introduced the concept of data warehouse to strengthen the rationalization of the system. When designing the data warehouse, we used traditional relation database theory to meet the demand of system of handling traditional management information system operations. On the other hand, we set up the fact and dimensionality table to meet the demand of analysis oriented. On the web structure design, the article selected the client/server structure, demonstrated the accuracy of the web structure, and configured the application environment of the server and the.The 4th part of the article offered introduction on the key technologies in implementation of this system. The first was the data preprocessing technology. The source data have different formats due to the variety of sources in the system. As our analysis was based on data, the data preprocessing technology became a difficulty and a key issue as well in implementation of the system. With preprocessing, data were standardized and ready to be analyzed. The second technology was OLAP. The functions of OLAP can be seen on two aspects. Firstly, it serves as the interface of new models in the system. We can save the data queried by OLAP in standard format and preset the interface to add new models easily or introduce the 3rd party software. Secondly, OLAP can systematically manage the maintenance of data dimensionality and keep the timing characteristics of the early warning analysis. As to the early warning method, ARCH, a currently widely used model was applied in the system. The article provided introduction on the structure, testing and parameter estimation of ARCH model. Setup of index system was also one of the key technologies in the implementation. By analyzing the characteristics of various indexes and discussing with experts from Sichuan Economy Committee, we used the industrial economy profit comprehensive indexas the index system. The structure of the industrial economy profit comprehensive index enabled its movement to be reflected by the index movement, which would be essential for analyzing and removing potential risks.The demonstration analysis in the article showed that the precision of Sichuan industrial economy early warning system was high and thus was able to provide conclusion with high value for reference. The industrial economy profit comprehensive index was proved to be helpful in analyzing and removing potential risks. The last part of the article pointed out the characteristics of the system and proposed some weaknesses to be updated in the future.