中文摘要：

为了度量金融市场的不确定性,本文引入了模糊变量。假设资产收益率为模糊数,分别运用可信性均值和可信性绝对偏差度量投资组合的收益与风险。考虑到投资者偏好,分别提出了以收益最大化的均值-绝对偏差优化模型和以风险最小化的优化模型。基于可信性理论,将上述模型转化为线性规划问题,并运用旋转算法求解。通过实证研究,证明了该算法的有效性,并比较了两个模型在实际投资决策过程中的效率。结果表明,以收益最大化的均值-绝对偏差优化模型效率优于风险最小的优化模型。

英文摘要：

Assuming that the return of asset is fuzzy number,the return and risk of portfolio in this paper are measured by the credibility mean and the credibility absolute deviation respectively.Thus two new credibilitic mean-absolute deviation fuzzy portfolio selection model are proposed considering the investor preferences.Based on the credibilitic theory,the proposed models are transformed into a crisp linear optimization problem.Then a pivoting algorithm is designed to obtain the optimal portfolio strategy.Finally,an example is given to illustrate the effectiveness of the proposed algorithm and test which model is more efficient in the decision-making process.