The Integration Modeling Method And Qualitative Simulation Research Of Macroeconomic Prediction System

With respect to complex system problem, traditional reductionism approach is usually ineffective owing to the complexity of related system, while qualitative-quantitative integration provides methodological guide for the modeling of complex system problem. For example, due to the complexity of economic system, macroeconomic prediction requires qualitative-quantitative integration in economic modeling in order to take comprehensive use of qualitative and quantitative knowledge and information. The research method of qualitative-quantitative integration adopted in macroeconomic prediction is a modeling metasynthesis method of human-computer combination. A problem concerned much by research faculty is that how the integration should be carried out. This paper makes a deep research on integration modeling of complex system problem in macroeconomic prediction.Based on relevant researches, this paper puts forward an integration modeling method based on dynamic model space and integration model.At first, we should notice that modeling method for complex system problem is scattered. A modeling method used in a model can usually be divided into several aspects. In this paper, the classification angle of modeling is called the attribute of modeling method, while the specific classifications are called the values of an attribute, and the value range of an attribute is called the state space of the predicable. With the modeling method as the research object, the attributes of modeling method are descriptive factors of the object, for which a factor space can be established. This factor space, which is a variable dimensional space, is called the attribute space of modeling method, i.e. the state space of all modeling method attributes.The variable dimensional structure of attribute space of modeling method acts on a model set, structuring a model space.A modeling method with the same attribute but different attribute values can take use of knowledge and information of different natures to build the model. In the modeling practice of complex system problem, a new integration method is always needed, which is to add new attribute value of integration method to...