COMPANY VALUATION AND PORTFOLIO ANALYSIS BASED ON K-MEANS CLUSTERING IN KOMPAS 100 STOCKS INDEX

The capital market plays a vital role in investment, providing a platform for trading long-term financial instruments. Indonesia’s capital market has shown significant growth in recent years. This study aims not only to find stock clusters but to show that grouping stocks based on similar valuation characteristics can serve as a solid foundation for constructing superior-performing portfolios. The Kompas 100 index is used because it represents the most liquid and fundamentally stocks in Indonesia. The k-means clustering method is employed, and the number of clusters is determined using the elbow method. This approach resulted in four clusters, with the cluster identified as containing stocks with low PER, PBV, and PSR, representing the “best” portfolio each year based on valuation. Portfolios were formed from these clusters and compared to benchmark portfolios in Indonesia and globally. Global portfolios used as benchmarks include VSMPX, FXAIX, and SAM Equity. Over five years (2018–2022), the cluster-based portfolios outperformed Indonesian and global benchmarks in 2018, 2021, and 2022, while slightly underperforming global portfolios in 2019 and 2020 but still exceeding Indonesian benchmarks. This confirms that clustering techniques can deliver strong performance compared to conventional methods. A limitation of this study is that it focuses only on return performance without analyzing risk-adjusted returns, which future research should address.