一般最小方差组合投资权系数(英文)

Generalized Minimum-Variance-Portfolio Weights

组合投资优化在组合投资管理中被广泛研究,在研究中,一般使用的是拉格朗日乘子法.然而,这一方法有某些限制:其基本假设是回报的方差阵是正定的,这使得该方法不能在一般情况下使用.本文作者的目标是应用二次优化理论以获得一般情况下的最优权系数,所得结果突破了前述的方差阵的限制.

Portfolio weights optimization has been extensively studied in the literature of portfolio management. The commonly used method is the Lagrange multiplier; however, this approach has some limitations:the fundamental assumption in this approach is that the covariance matrix of returns is positive definite, which renders the method not applicable in general. In this paper, the authors aim to use quadratic optimization theory in obtaining generalized optimal weights, whereby, the restriction on the covariance matrix is just a mere special case.