Anugrahayu, Mella Refina (2023) Optimasi Portofolio Saham Menggunakan Model Mean-Variance dan Mean Absolute Deviation Berdasarkan K-Medoids Clustering dengan Pendekatan Dynamic Time Warping. Other thesis, Institut Teknologi Sepuluh Nopember.
Text
06311940000020-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (3MB) | Request a copy |
Abstract
Kecenderungan investor untuk memilih investasi dengan return maksimal dan risiko minimal mengakibatkan perlunya diversifikasi dalam suatu portofolio untuk membentuk portofolio optimal. Salah satu alternatif optimasi portofolio dapat dilakukan menggunakan analisis pengelompokan (clustering). Namun, clustering hanya terbatas untuk menentukan kandidat saham optimal, sehingga perlu ditambah metode atau model optimasi lain untuk menghitung bobot portofolio. Model pembentukan portofolio optimal seperti model Mean-Variance (MV) dan Mean Absolute Deviation (MAD) menggunakan asumsi bahwa preferensi investor didasarkan pada tingkat expected return dan risiko dari portofolio, tetapi cara memilih saham untuk model tersebut tidak didiversifikasi secara detail, sehingga dalam penelitian ini, dilakukan penggabungan metode pembentukan portofolio optimal antara model optimasi MV dan MAD dengan analisis pengelompokan (clustering) saham menggunakan metode K-Medoids Clustering dengan pendekatan ukuran jarak Dynamic Time Warping (DTW). Idealnya, dalam pembentukan portofolio optimal juga disertai dengan perhitungan estimasi risiko yang akan diperoleh investor. Alternatif pengestimasian risiko yang digunakan dalam penelitian ini adalah metode Expected Tail Loss (ETL) berdasarkan hasil Simulasi Monte Carlo. Variabel yang digunakan dalam penelitian ini adalah data saham yang konsisten terdaftar dalam Indeks SRI-KEHATI periode 1 Mei 2017 hingga 31 Desember 2022 dan tingkat suku bunga IndONIA sebagai aset bebas risiko (risk free rate). Berdasarkan analisis yang telah dilakukan, portofolio MAD merupakan portofolio yang lebih optimal dibandingkan portofolio MV dengan portofolio MAD yang tersusun atas lima saham yaitu saham BMRI dengan bobot sebesar 0,06243, saham UNTR sebesar 0,08658, saham BBRI sebesar 0,10285, saham BBCA sebesar 0,53623, dan saham KLBF sebesar 0,21191 menjadi portofolio optimal yang terbaik. Portofolio optimal model MAD memiliki tingkat pengembalian (return) sebesar 87,836% dalam kurun waktu Mei 2017 – Desember 2022 dengan kinerja portofolio sebesar 0,03704, sedangkan tingkat risiko yang dihasilkan berdasarkan Monte Carlo-Expected Tail Loss adalah sebesar 2,2416%.
=================================================================================================================================
The tendency of investors to choose investments with maximum return and minimal risk results in the need for diversification in a portfolio to form an optimal portfolio. One alternative for portfolio optimization is using clustering analysis. However, clustering is limited to determining optimal stock candidates, so additional methods or optimization models are needed to calculate portfolio weights. Optimal portfolio construction models such as Mean-Variance (MV) and Mean Absolute Deviation (MAD) assume that investor preferences are based on the expected return and risk level of the portfolio, but the selection of stocks for these models is not extensively diversified. Therefore, in this study, a combination of optimal portfolio construction methods between the MV and MAD optimization models with stock clustering analysis using the K-Medoids Clustering method and the Dynamic Time Warping (DTW) distance measurement approach is performed. Ideally, in the formation of an optimal portfolio, the estimation of the risk to be obtained by the investor is also included. The alternative risk estimation method used in this study is the Expected Tail Loss (ETL) method based on Monte Carlo Simulation results. The variables used in this study are consistent stock data listed in the SRI-KEHATI Index from May 1, 2017, to December 31, 2022, and the IndONIA interest rate as the risk-free asset (risk-free rate). Based on the analysis conducted, the MAD portfolio is more optimal than the MV portfolio, with the MAD portfolio consisting of five stocks, namely BMRI shares with a weight of 0,06243, UNTR shares of 0,08658, BBRI shares of 0,10285, BBCA shares of 0,53623, and KLBF shares of 0,21191 being the best optimal portfolio. The optimal portfolio of the MAD model has a return rate of 87,836% over the period from May 2017 to December 2022, with a portfolio performance of 0,03704. However, the risk level based on Monte Carlo-Expected Tail Loss is 2,2416%.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | Dynamic Time Warping, Mean Absolute Deviation, Monte Carlo, Expected Tail Loss¸ K-Medoids Clustering, Mean-Variance |
Subjects: | H Social Sciences > HG Finance > HG4529 Investment analysis H Social Sciences > HG Finance > HG4529.5 Portfolio management Q Science > QA Mathematics > QA278.55 Cluster analysis |
Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis |
Depositing User: | Mella Refina Anugrahayu |
Date Deposited: | 12 Jul 2023 07:48 |
Last Modified: | 12 Jul 2023 07:48 |
URI: | http://repository.its.ac.id/id/eprint/98429 |
Actions (login required)
View Item |