SYSTEM AND METHOD FOR OPTIMIZING PORTFOLIO AND EXECUTABLE DIGITAL MEDIA STORING COMPUTATIONAL PROGRAMS IMPLEMENTING THE SAME

摘要

The present invention relates to a portfolio optimization system and method. The portfolio optimization system of the present invention can be used to measure the loss risk of a portfolio or portfolio constituent asset group by using an indicator such as conditional loss VaR having a segmented linearity with respect to individual asset group value ratios constituting the portfolio, Optimization under constraints on total asset group adjustment cost, total portfolio budget, total portfolio loss risk index, or total portfolio profit index, using time series matrices representing the change in value of asset groups and profit indicators of asset groups as input values. State Portfolio State Indicators are implemented and output with a second order nonlinear approximation algorithm. When the total number N of asset classes constituting the portfolio is any natural number greater than or equal to 2, the portfolio initial and optimal states correspond to the points in the N-dimensional space, and in the direction toward the optimization path connecting the points successively in the N- A parameter space consisting of a maximum of three constraints, which necessarily includes the entire asset group adjustment cost within a local trace range that takes the form of a very small N-dimensional ellipsoidal boundary formed around a point on the path, Dimensional nonlinear projection transformation to the low dimensional nonlinear projection transformation, which is derived from the Lagrangian multiplication method for the portfolio optimization. According to the present invention, it is possible to calculate a segmented continuous path in which the objective function of the current portfolio changes to the optimized state.