| Abstrakt: |
Optimization theory in finance is developing rapidly, especially in forming portfolio optimization by including fuzzy elements. However, some fuzzy methods encountered problems in their application when obtaining optimal portfolio weights. Hence, a multi-objective approach was required to solve the fuzzy portfolio. This research focused on compiling a fuzzy portfolio using large-scale data without considering short-selling. Then, the data consisting of opening, closing, highest, and lowest prices were modeled into non-linear adaptive fuzzy numbers. The incoming fuzzy number was a trapezoidal fuzzy number. The data was obtained from 491 trading days of the ten most active stocks in LQ45, respectively. This research also employed the Treynor ratio (TR) to optimize the process. As a control for the Treynor ratio, the Sharpe ratio (SR), which had been employed, was also first introduced. Based on empirical data, there were differences in the composition of the weights resulting from the formed fuzzy portfolio. The research results significantly indicated that TR had a more even diversification level across all stocks when constructing a fuzzy portfolio with a multi-objective approach. It indeed reduced the systematic risk in the portfolio formed. [ABSTRACT FROM AUTHOR] |
