Pembentukan Portofolio Optimal Saham-Saham LQ45 Berdasarkan Sentimen Positif Menggunakan Metode DBSCAN dan Markowitz

Suputra, I Gede Dana (2025) Pembentukan Portofolio Optimal Saham-Saham LQ45 Berdasarkan Sentimen Positif Menggunakan Metode DBSCAN dan Markowitz. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2043211010-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (3MB) | Request a copy

Abstract

Perkembangan pasar modal di Indonesia dapat dilihat dari meningkatnya jumlah investor dan jumlah perusahaan yang terdaftar di Bursa Efek Indonesia (BEI). Dalam berinvestasi, investor perlu memperhatikan berbagai informasi untuk memilih saham terbaik, salah satunya berita. Melalui berita akan muncul pandangan masyarakat terhadap saham perusahaan terkait berupa pandangan positif, negatif, atau netral yang diperoleh menggunakan analisis sentimen. Media sosial X dapat menjadi sumber informasi bagi investor untuk memperoleh sentimen saham perusahaan. Berinvestasi saham merupakan kegiatan yang memiliki risiko tinggi, sehingga penentuan portofolio optimal perlu dilakukan agar tujuan investasi dapat tercapai. Terdapat banyak cara untuk membentuk portofolio optimal, diantaranya dengan menggabungkan metode clustering DBSCAN dan metode Markowitz. Metode DBSCAN digunakan untuk mengelompokkan saham berdasarkan rasio keuangan dan sentimen saham. Kelompok-kelompok yang terbentuk akan ditentukan portofolio optimalnya menggunakan metode Markowitz. Berdasarkan hasil analisis, menggunakan analisis sentimen diketahui persentase sentimen positif terbesar adalah 76,47% dan terkecil 45,04% dengan rata-rata 64,05%. Dilanjutkan dengan mengelompokkan saham-saham tersebut berdasarkan ROA, ROE, EPS, dan sentimen positif menggunakan DBSCAN, terbentuk 3 cluster dan 1 cluster noise. Selanjutnya akan dibentuk portofolio optimal dari setiap cluster menggunakan model Markowitz. Pada cluster 1 terdapat 10 saham penyusun portofolio optimal. Pada cluster 2 terdapat 8 saham penyusun portofolio optimal. Pada cluster 3 terdapat 4 saham penyusun portofolio optimal. Pada cluster noise terdapat 12 saham penyusun portofolio optimal. Keempat portofolio tersebut memiliki tingkat risiko yang berbeda-beda. Cluster noise merupakan cluster yang memiliki portofolio paling optimal karena dengan return yang sama, cluster noise memiliki risiko paling rendah.
=================================================================================================================================
The development of the capital market in Indonesia can be observed from the increasing number of investors and companies listed on the Indonesia Stock Exchange (IDX). When investing, investors need to consider various information to select the best stocks, one of which is news. News provides public sentiment regarding a company's stocks, such as positive, negative, or neutral sentiments, obtained through sentiment analysis. Social media platform X can serve as a source of information for investors to gather stock sentiment. Stock investing is a high-risk activity, making it essential to determine an optimal portfolio to achieve investment goals. There are many ways to form an optimal portfolio, including combining the DBSCAN clustering method with the Markowitz model. The DBSCAN method is used to cluster stocks based on financial ratios and stock sentiment. The clusters formed are then used to determine the optimal portfolio using the Markowitz model. Based on the analysis, sentiment analysis results show that the highest positive sentiment percentage is 76.47%, and the lowest is 45.04%, with an average of 64.05%. The stocks are then clustered based on ROA, ROE, EPS, and positive sentiment using DBSCAN, resulting in 3 clusters and 1 noise cluster. Subsequently, optimal portfolios are formed for each cluster using the Markowitz model. Cluster 1 contains 10 stocks, Cluster 2 includes 8 stocks, Cluster 3 has 4 stocks, and the noise cluster consists of 12 stocks. Each portfolio has different risk levels, with the noise cluster being the most optimal as it offers the lowest risk while maintaining the same return.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Analisis Sentimen, DBSCAN, Markowitz, Portofolio, Saham. Sentimen Analysis, DBSCAN, Markowitz, Portfolio, Stocks.
Subjects:	Q Science > QA Mathematics > QA278.55 Cluster analysis Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics) T Technology > T Technology (General) > T174.5 Technology--Risk assessment. T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T57.62 Simulation
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	I Gede Dana Suputra
Date Deposited:	17 Feb 2025 00:36
Last Modified:	17 Feb 2025 00:36
URI:	http://repository.its.ac.id/id/eprint/118758

Actions (login required)

View Item