Pengelompokan Saham Sektor Keuangan Dengan Metode K-Means Clustering With Dynamic Time Warping Distance

Aqsari, Hasri Wiji (2023) Pengelompokan Saham Sektor Keuangan Dengan Metode K-Means Clustering With Dynamic Time Warping Distance. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6003202019-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (5MB) | Request a copy

Abstract

Investor adalah orang yang menanamkan uangnya dalam usaha dengan tujuan mendapatkan keuntungan. Instrumen investasi di pasar modal diantaranya adalah saham, obligasi, waran, right, reksa dana dan berbagai instrumen derivatif lainnya. Jumlah investor saham mengalami peningkatan dalam satu tahun ini, dimana sektor saham keuangan termasuk yang tumbuh pesat. Setiap investor tentunya ingin mendapatkan keuntungan dari saham yang dimilikinya. Sehingga perlu dipertimbangkan kelompok saham mana saja yang memiliki fluktuasi harga yang baik. Pada penelitian ini menggunakan data harga saham perusahaan sektor keuangan pada periode 01 April 2021 sampai dengan 31 Maret 2022. Variabel yang digunakan adalah harga saham open, close dan HML (High Minus Low). Metode yang digunakan adalah K-Means clustering. Jarak yang digunakan adalah Dynamic Time Warping (DTW). DTW dipilih karena merupakan jarak non-linear sequence alignment sehingga dinilai cocok untuk diaplikasikan ke dalam data saham yang berbentuk data time series. Terakhir akan dibandingkan hasil dari pengelompokan metode K-Means menggunakan jarak Euclidean dan Dynamic Time Warping (DTW). Perbandingan keduanya menggunakan nilai Silhouette Score. Hasil analisis data simulasi adalah performa jarak DTW lebih baik dibandingkan dengan jarak euclidean ketika plot data terjadi overlapping. Hasil analisis data saham sektor keuangan adalah harga saham open dan close terbagi menjadi dua kelompok yaitu kelompok pertama saham perusahaan besar dan kelompok kedua adalah saham perusahaan kecil. Sedangkan untuk harga saham HML terdapat sedikit perbedaan anggota dengan saham open dan close
=====================================================================================================================================
An investor is a person who invests money in a business with the aim of making a profit. Investment instruments in the capital market include stocks, bonds, warrants, rights, mutual funds and various other derivative instruments. The number of stock investors has increased in this one year, where the financial stock sector is one of the fastest growing. Every investor wants to benefit from the shares they own. So it is necessary to consider which groups have good price fluctuations. In this study, data on share prices of financial sector companies are used for the period 01 April 2021 to 31 March 2022. The variables used are open, close and HML (High Minus Low) stock prices. The method used is K-Means clustering. The distance used is Dynamic Time Warping (DTW). DTW was chosen because it is a non-linear sequence alignment distance, so it is considered suitable to be applied to stock data in the form of time series data. Finally, the results of the K-Means method grouping will be compared using Euclidean distance and Dynamic Time Warping (DTW). Comparison of the two using the Silhouette Score. The result of the simulation data analysis is that the performance of DTW distances is better than the euclidean distance when the data plots overlap. The result of financial sector stock data analysis is that the open and close stock prices are divided into two groups: the first group is shares of large companies and the second group is shares of small companies. As for the price of HML stocks, there is a slight difference between members with open and closed stocks

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Euclidean, Dynamic Time Warping, K-Means, Saham, Silhouette Score
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA278.55 Cluster analysis Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49101-(S2) Master Thesis
Depositing User:	Hasri Wiji Aqsari
Date Deposited:	17 Feb 2023 02:58
Last Modified:	17 Feb 2023 02:58
URI:	http://repository.its.ac.id/id/eprint/97506

Actions (login required)

View Item