Inferensi
Vol 9 No 2 (2026)

Stock Portfolio Optimization Based on Financial and Risk–Return Clustering and TOPSIS with MVEP–MAD Weighting

Wirda Andani (Statistics Study Program, Universitas Tanjungpura, Pontianak, Indonesia)
Shantika Martha (Statistics Study Program, Universitas Tanjungpura, Pontianak, Indonesia)
Evy Sulistianingsih (Statistics Study Program, Universitas Tanjungpura, Pontianak, Indonesia)
Muhammad Fikri (Statistics Study Program, Universitas Tanjungpura, Pontianak, Indonesia)
Cinta Priscillia Maharani (Statistics Study Program, Universitas Tanjungpura, Pontianak, Indonesia)
Rifki Pebriyandi (Statistics Study Program, Universitas Tanjungpura, Pontianak, Indonesia)



Article Info

Publish Date
26 Aug 2026

Abstract

This research aims to construct an optimal stock portfolio from the Kompas100 index using stock performance indicators, fundamental indicators, K-Means, TOPSIS, and portfolio optimization. Of the 100 stocks, only 22 were suitable as candidates for portfolio formation. From these 22 stocks, 4 portfolio candidates were identified through K-means analysis and 7 through TOPSIS analysis. The next step was to determine the investment proportion for each stock in the portfolio using MVEP and MAD. Performance evaluation results show that Portfolio 4, consisting of PTRO and WIFI stocks, consistently yields the highest Sharpe Ratio under both weighting methods: 0.19 using MVEP and 0.21 using MAD. Portfolio 4’s performance was then re-evaluated using data from April through December 2025, resulting in a higher Sharpe ratio for both the MAD and MVEP. Overall, this study demonstrates that the combination of the K-Means Clustering, TOPSIS, MVEP, and MAD methods can be used to assist in the stock selection process and the formation of an optimal portfolio that is more efficient than investing in a single stock because it provides a better balance between return and risk through investment diversification, while remaining stable for the next nine months.

Copyrights © 2026






Journal Info

Abbrev

inferensi

Publisher

Subject

Computer Science & IT Decision Sciences, Operations Research & Management Engineering Mathematics Social Sciences

Description

The aim of Inferensi is to publish original articles concerning statistical theories and novel applications in diverse research fields related to statistics and data science. The objective of papers should be to contribute to the understanding of the statistical methodology and/or to develop and ...