Optimisasi Portofolio Saham IDX30 Menggunakan K-Prototype Clustering Dan Mean-Conditional Value At Risk

Shrinishaa, Rajni (2025) Optimisasi Portofolio Saham IDX30 Menggunakan K-Prototype Clustering Dan Mean-Conditional Value At Risk. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5006211082-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Dalam era globalisasi dan perkembangan teknologi yang pesat, pasar modal menjadi semakin kompleks dan menunutut strategi investasi yang lebih cermat. Investor perlu mengelola portofolio secara efektif untuk memaksimalkan imbal hasil sekaligus meminimalkan risiko. Penelitian ini akan membahas pembentukan portofolio optimal dengan estimasi bobot dan risiko melalui pendekatan optimisasi Mean-CVaR dari portofolio yang terbentuk dari hasil pengelompokkan menggunakan K-Prototype Clustering dan metode Elbow untuk penentuan jumlah klaster optimal. Data saham dalam penelitian ini adalah data price 27 saham yang terdaftar di Indeks IDX30 selama periode Januari 2020 hingga November 2024. Pengelompokkan saham dilakukan dengan pertimbangan kriteria berupa volatilitas, volume, ROA, ROE, EPS, kapitalisasi relatif, dan sektor saham. Hasil penelitian menunjukkan bahwa saham dapat dikelompokkan menjadi empat dan lima cluster dengan kriteria yang telah ditetapkan kecuali nilai ROA yang memberikan korelasi yang tinggi pada variabel lainnya. Pembentukan portofolio menggunakan Sharpe Ratio menghasilkan dua potofolio dari hasil nilai rasio tertinggi tiap klaster, dimana portofolio pertama berisi saham MDKA, AKRA,ARTO, dan ANTM, sedangkan portofolio kedua terdiri dari saham BBCA, AMRT, ANTM, ARTO, dan MDKA. Berdasarkan pendekatan Mean-CVaR, portofolio pertama cenderung memberikan return yang lebih tinggi, namun dengan risiko ekstrem yang lebih besar, sedangkan portofolio kedua menawarkan kestabilan risiko meski dengan return yang lebih rendah. Evaluasi menggunakan Sharpe Ratio menunjukkan bahwa portofolio pertama lebih optimal bagi investor yang toleransi risiko rendah hingga menengah, sementara portofolio kedua lebih sesuai untuk investor denagn toleransi risiko tinggi yang mengutamakan keuntungan.
======================================================================================================================================
In the era of globalization and rapid technological development, the capital market is becoming increasingly complex and requires a more careful investment strategy. Investors need to manage portfolios effectively to maximize returns while minimizing risk. This study will discuss the formation of an optimal portfolio with risk estimation through the Mean-CVaR optimization approach from a portfolio formed from the results of grouping using K-Prototype Clustering and the Elbow method to determine the optimal number of clusters. The stock data in this study are price data of 27 stocks listed on the IDX30 Index during the period January 2020 to November 2024. Stock grouping is carried out by considering criteria such as volatility, volume, ROA, ROE, EPS, relative capitalization, and stock sector. The results of the study show that stocks can be divided into four and five clusters with predetermined criteria except for the ROA value which provides a high correlation with other variables. Portfolio formation using the Sharpe Ratio produces two portfolios from the highest ratio values of each cluster, where the first portfolio contains MDKA, AKRA, ARTO, and ANTM stocks, while the second portfolio consists of BBCA, AMRT, ANTM, ARTO, and MDKA stocks. Based on the Mean-CVaR approach, the first portfolio tends to provide higher returns, but with greater extreme risks, while the second portfolio offers risk stability even with lower returns. Evaluation using the Sharpe Ratio shows that the first portfolio is more optimal for investors with low to medium risk tolerance, while the second portfolio is more suitable for investors with high risk tolerance who prioritize profits.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Conditional Value at Risk, K-Prototype Clustering, , Mean-Conditional Value at Risk, Metode Elbow, Optimisasi Portofolio, Optimization Portfolio, Elbow Method, K-Prototype Clustering, Conditional Value at Risk, Mean-Conditional Value at Risk
Subjects:	H Social Sciences > HG Finance > HG4529.5 Portfolio management H Social Sciences > HG Finance > HG4915 Stocks--Prices Q Science > QA Mathematics > QA278.55 Cluster analysis
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Rajni Shrinishaa
Date Deposited:	31 Jul 2025 08:42
Last Modified:	31 Jul 2025 08:42
URI:	http://repository.its.ac.id/id/eprint/125173

Actions (login required)

View Item