Research on Methods of Fuzzy Portfolio Selection with Limited Number of Securities
Grant-in-Aid for Scientific Research (C)
|Allocation Type||Single-year Grants|
|Research Institution||Osaka Institute of Technology|
WATADA Junzo Osaka Institute of Technology Faculty of Engineering.Professor, 工学部, 教授 (10158610)
|Project Period (FY)
1996 – 1998
Completed(Fiscal Year 1998)
|Budget Amount *help
¥2,200,000 (Direct Cost : ¥2,200,000)
Fiscal Year 1998 : ¥600,000 (Direct Cost : ¥600,000)
Fiscal Year 1997 : ¥500,000 (Direct Cost : ¥500,000)
Fiscal Year 1996 : ¥1,100,000 (Direct Cost : ¥1,100,000)
|Keywords||Fuzzy portfolio selection / Fuzzy mathematical programming / Fuzzy quadratic programming / Genetic algorithm / Mean-variance analysis / Boltzmann Machine / Hierarchical Boltzmann Machine / Selection into a limited number / ファジィ理論 / 希求水準 / ニューラルネットワーク|
This research has been pursued and accomplished for three years. We have worked on developing the solutions of problems cleared and remained in the two previous researches and have obtained the following results.
1)The interviews with experts on investment in the first and second research years cleared that it is important to reflect the judgment obtained from experts on investment on the decision-making.
2)Especially, we successfully imprinented expectation level of invest experts in the model of fuzzy portfolio selection.
3)As the main objective of this research, we developed the fuzzy model that can enable us to select the limited number of stocks out of huge stock market as the result.
4)As dealing of stocks easily disturbs the stock market and we should take the dealing fee into consideration. it is preferred to select portfolio pattern that has minimum changes from the previous portfolio pattern. We proposed the model that enables us to select portfolio with minimum changes.
5)In the Boltzmann Machine Model of portfolio selection, the adjustment of the weight between expected return and risk can be changeable according preference of strategy with high risk and high expected return to strategy with low risk and low expected return.
6)Furthermore, we modeled the hierarchical Boltzmann Machine that enables us to select the limited number of stocks. This model will be presented at the international conference of fuzzy system in August 1999.
7)We also obtained the following results by developing the above-mentioned research. We have also bui It the approach of the Boltzmann Machine's to solving effectively and efficiently a portfolio select ion problem.
8)As these applications, we have modeled a hierarchical investment strategy by a Boltzmann Machine to invest.money in the international stock market.
Research Output (27results)