What does this research mean for the field?

Applying the CLARA clustering method and Mean-Variance Efficient Portfolio weighting to IDX80 stocks identifies an optimal four-stock diversified portfolio with a maximum daily Value at Risk of approximately 1.37% at a 99% confidence level. Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

The study aims to construct an optimal investment portfolio using clustering methods and assess risk through Value at Risk analysis.

April 12, 2026SHILAP Revista de lepidopterología0 citationsOpen Access

Constructing an Optimal Portfolio Using Clustering Large Application and Value at Risk Analysis for Idx80 Stocks

SPSania PujiantiTanjungpura University HPHendra PerdanaTanjungpura University NSNeva SatyahadewiTanjungpura University

Key Points

The study aims to construct an optimal investment portfolio using clustering methods and assess risk through Value at Risk analysis.
Utilized Clustering Large Application (CLARA) to group IDX80 stocks based on performance.
Conducted a risk analysis using Value at Risk (VaR) to estimate potential losses.
Selected representative stocks from clusters based on the highest Sharpe ratio.
Identified four stock clusters with a low silhouette coefficient of 0.18226.
Selected stocks for the optimal portfolio: SCMA, JPFA, GOTO, and BRIS, with specific weights.
Calculated a potential maximum loss of Rp137,139 at a 99% confidence level for an initial investment of Rp10,000,000.

Abstract

Investment is a way to manage wealth and achieve financial goals in the future. Stocks are an attractive investment instrument due to their high potential returns, although they also carry significant risks. These risks can be minimized through portfolio diversification. Diversification is carried out by selecting representative stocks from the clustering results. This study aims to construct an optimal portfolio using the Clustering Large Application (CLARA) method and conduct portfolio risk analysis using Value at Risk (VaR). The data used includes IDX80 stock closing prices from November 1, 2024, to January 31, 2025, the financial ratios of IDX80 stocks on December 2024, and the Bank Indonesia (BI-Rate) interest rate from November 2024 to January 2025. The CLARA method produces four stock clusters with a silhouette coefficient of 0.18226. This value indicates a low level of separation between clusters, as there might be overlapping features among the clusters. Representative stocks from each cluster are selected based on the highest Sharpe ratio: SCMA, JPFA, GOTO, and BRIS. The portfolio weights based on MVEP are 15.002% (SCMA), 29.786% (JPFA), 1.858% (GOTO), and 53.354% (BRIS). The VaR calculation shows a potential maximum loss of Rp137,139 in one day, with a 99% confidence level, from an initial investment of Rp10,000,000.

AI에게 질문

Bookmark

View Full Paper