PENGELOMPOKAN SAHAM MENGGUNAKAN K-MEANS DALAM PEMBENTUKAN PORTOFOLIO OPTIMAL

Abstract

K-Means clustering analysis is a technique used in grouping objects that have similar characteristics. In forming a portfolio, investors need a group of stocks from different sectors that aim to build a well-diversified portfolio. Portfolio diversification is the placement of assets from various stocks in such a way that risks can be minimized. This study aims to obtain the results of grouping stocks with K-Means at IDX80 and then determine the optimal portfolio of each cluster formed using the Mean Variance method in the period January, 1^st 2020 to November, 10^th 2022. As a result, obtained in this study that grouping with K -Means produces four groups and  is the best portfolio consisting of 10 stocks with a Sharp ratio performance value of 0.0062 with a risk portfolio of 1.59% and an expected return portfolio of 0.17%.