An application of artificial intelligence techniques to describing economic states and simulating economic evaluations
| Project/Area Number |
62450072
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
経済事情及び政策学
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
TSUKUI Jinkichi Professor of the Institute of Social and Economic Research, Osaka Unv., 社会経済研究所, 教授 (00029919)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SATO Hideto Associate Professor of the Institute of Social and Economic Research, Osaka Univ, 社会経済研究所, 助教授 (20114085)
YOSHIKAWA Hiroshi Associate Professor of the Institute of Social and Economic Research, Osaka Univ, 社会経済研究所・東京大学経済学部, 助教授 (30158414)
BAN Kanemi Associate Professor of the Faculty of Economics, Osaka University, 経済学部, 助教授 (30027578)
ONIKI Hajime Professor of the Institute of Social and Economic Research, Osaka University, 社会経済研究所, 教授 (40107107)
|
|---|
| Project Period (FY) |
1987 – 1988
|
| Project Status |
Completed (Fiscal Year 1988)
|
| Budget Amount *help |
¥3,500,000 (Direct Cost: ¥3,500,000)
Fiscal Year 1988: ¥1,500,000 (Direct Cost: ¥1,500,000)
Fiscal Year 1987: ¥2,000,000 (Direct Cost: ¥2,000,000)
|
|---|
| Keywords | economic evaluation / business judgement / qualitative data / expert system / artificial intelligence / inductive inference / automatic rule induction / 経済分析方法論 / 統計手法 / 数量化2類 / 景気診断 / シミュレーション / エキスパート / システム / 知識ベース / 判断知識 |
|---|
| Research Abstract |
1. Critical study of the inductive inference technique and an attempt of its application for the formulation of evaluation rules of economic states.
For establishing an expert system, it is usually done to extract Knowledge of experts through interview. Efficient application of this method is, however, difficult for economic problems since the interview data only relate to a particular moment of ever changing economic states. Therefore, we tried here another technigue which was called as the rule induction method. This method is recently attrcting researcher's attention in the field of A.I. We, first, investigated the relations between this method and the statistical inference which has been a typical analytical method in economics. We found that these two methods were complementary for each other. We, then, formulated the ecaluation rules of economic states based on the rule induction method by using the qualitative data mentioned at 3.
Evaluation of A.I. techniques for applications in economics.
By investigating methodlogical limitation of the present econometric model analysis, we showed usefulness and applicability of A.I. techniques for economic analysis.
3. Collection and systematization of qualitative data of economic state evaluation.
For the investigation of actual processes of economic evaluation, it is indispensable to collect the qualitative information about such processes and the data utilized in evaluation. In view of availability of such information we decided to utilize the Monthly Economic Reports of the Japanese government as basic data. We, then, built up the time series qualitative economic data based on the statements and the indices in these reports which covered 27 periods in the recent business cycles. In contray with quantitative data our data contain information of distributions and trends about economic states which will be useful for the attempt of simulating human ecaluation process.
|
Report
(3 results)
Research Products
(7 results)
