| Budget Amount *help |
¥1,200,000 (Direct Cost: ¥1,200,000)
Fiscal Year 2006: ¥500,000 (Direct Cost: ¥500,000)
Fiscal Year 2005: ¥700,000 (Direct Cost: ¥700,000)
|
|---|
| Research Abstract |
This research objective is to define Economic Stabilization Policy as Optimal Stabilization policy and Phillips-type Feedback Control policy, organize the dynamic structure and exercise the policy simulation. In facing difficulty of modeling by quantization for policy model, fuzzy control, neural network algorithm can be utilized by combining these two technologies for measuring the policy effects. This research pursues for the availability to use the existing statistical and econometric software for this policy approach. Having compared with these packages can be used for estimating the structural equations and policy simulations, but cannot be used for fuzzy control and neural network algorithm. MATLAB is only suitable package for fuzzy control policy, then also available for economic stabilization policy. In particular, it is understandable that the Phillips-PID-type control policy can be complemented by fuzzy control. Conventionally, stabilization policies have nonlinear constraints so that there is difficulty to control numerically., but it is a great contribution that this neuro-fuzzy approach enable to adopt nonlinear constraints, even policy language relations. Furthermore, for function of packages, it can be understandable that the programming language package is more easy and universal to use for formulation of policy setting. In the future assignment, there is some room to consider how effective the neuro-fuzzy policy approach can be changed by the kind of various econometric computations. There remains how effective this approach is for forecasting and controlling the large-scale economic and corporate systems by modeling using the actual data
|
