| Budget Amount *help |
¥3,120,000 (Direct Cost: ¥2,400,000、Indirect Cost: ¥720,000)
Fiscal Year 2012: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2011: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
|
|---|
| Research Abstract |
The following three themes studies were executed during these two years.
(1)Forecasting accuracy improvement
(2)Analysis of the matrix structure in brand selection
(3)Improving the machine diagnosis techniques
All these are the applications of Time Series Analysis.
(1)A new trend removal method by utilizing Genetic Algorithm and a new forecasting method by utilizing Neural Network were devised.
(2)A new analysis model of the brand shift from maker to maker, the extension of the model to the second order lag and the model involving customer satisfaction were devised.
(3)The phenomenon that Kurtosis went up and down as the deterioration proceeded was traced and its reason was clarified by the analysis
Accuracy and meaning of the proposed methods were verified on the real data such as the food shipping data, the jewelry purchasing data and theautomobile purchasing data.
|
