首頁 > 網路資源 > 大同大學數位論文系統

Title page for etd-0216117-205455


URN etd-0216117-205455 Statistics This thesis had been viewed 579 times. Download 0 times.
Author Bo-han Chen
Author's Email Address No Public.
Department Information Management
Year 2016 Semester 1
Degree Master Type of Document Master's Thesis
Language zh-TW.Big5 Chinese Page Count 54
Title USING CUCKOO OPTIMIZATION ALGORITHM TO SOLVE PORTFOLIO OPTIMIZATION PROMBLEMS
Keyword
  • Portfolio Optimization
  • Efficient frontier
  • Cuckoo optimization algorithm
  • Cuckoo optimization algorithm
  • Efficient frontier
  • Portfolio Optimization
  • Abstract In Portfolio optimization (PO) problem, we need to consider about the set of assets that we want to invest and allocate the fund at the same time. PO is a multi-Object combinatorial optimization problem and the goal is to maximize the expected return and to minimize the risk. The problem becomes more complex when the problem size (number of assets) is large. Cuckoo optimization algorithm (COA) is inspired by the way of breeding and immigration of the cuckoos in the real world. COA is suitable for solving continuous nonlinear optimization problems. Due to its fast convergence, COA can reach global optima in fewer iterations. This Study tries to introduce the new algorithm to solve PO problem. In egg-laying, every cuckoo solution evolves the proportion of a stock in order to improve its objective function value for the set of selected stocks. Using immigration, cuckoo solutions learn from the current "goal point" so that cuckoos can gradually converge into same area. The proposed algorithm was tested on 5 different sized of problems for the quality and effectiveness, moreover, the results are compared with existing algorithms. The experimental results show that COA has decent performance for solving PO problem.
    Advisor Committee
  • Yucheng Kao - advisor
  • Wen-Hwa Liao - co-chair
  • Yun-Chia Liang - co-chair
  • Files indicate in-campus access at 5 years and off-campus not accessible
    Date of Defense 2016-07-20 Date of Submission 2017-02-17


    Browse | Search All Available ETDs