Clustering LQ45 Stocks Using Dynamic Time Warping and K-Medoids
Stock closing prices form dynamic, nonlinear time series in which similar patterns may occur at different times. This study groups LQ45 stocks using Dynamic Time Warping (DTW) and K-Medoids. DTW measures dissimilarity between price series by allowing nonlinear temporal alignment, while K-Medoids forms clusters from the resulting dissimilarity matrix. The dataset comprises daily closing prices of 45 LQ45 stocks from 1 August to 30 November 2025, obtained from Yahoo Finance. The analysis includes data completeness checks, DTW distance calculation, K-Medoids clustering, and evaluation using the Davies Bouldin Index (DBI). Among the cluster counts tested, from two to nine, three clusters yield the lowest DBI of 0.24193. The clusters contain seven, 36, and two stocks, with INKP, ISAT, and UNTR as their respective medoids. The clusters differ in their price movement patterns and magnitudes of price fluctuations. These results indicate that DTW and K-Medoids can group LQ45 stocks according to similarities in closing price movements during the study period.
Authors
- Moch. Abdul Mukid (ORCID: https://orcid.org/0000-0002-9252-4949)
- Agus Rusgiyono
- Yohanes Abel Febi Christian
Institutions
- Diponegoro University (ID)
Publication Details
- Journal
- INTERNATIONAL JOURNAL OF MATHEMATICS AND COMPUTER RESEARCH
- Published
- 2026-11-18
- DOI
- https://doi.org/10.5281/zenodo.24363348
- Primary Topic
- Time Series Analysis and Forecasting
- Type
- article
- Field-Weighted Citation Impact
- 0.00