PREDIKSI VOLATILITAS HARGA SAHAM MENGGUNAKAN MODEL MULTIVARIAT GARCH PADA INDUSTRI KOSMETIK INDONESIA

Ikhsan, Renaldy Al (2025) PREDIKSI VOLATILITAS HARGA SAHAM MENGGUNAKAN MODEL MULTIVARIAT GARCH PADA INDUSTRI KOSMETIK INDONESIA. Undergraduate thesis, Universitas Pembangunan Nasional Veteran Jawa Timur..

[img] Text (Cover)
21083010072_Cover.pdf

Download (1MB)
[img] Text (Bab 1)
21083010072_BAB I.pdf

Download (288kB)
[img] Text (Bab 2)
21083010072_BAB II.pdf
Restricted to Repository staff only until 25 July 2028.

Download (901kB)
[img] Text (Bab 4)
21083010072_BAB III.pdf
Restricted to Repository staff only until 25 July 2028.

Download (463kB)
[img] Text (Bab 4)
21083010072_BAB IV.pdf
Restricted to Repository staff only until 25 July 2028.

Download (6MB)
[img] Text (Bab 5)
21083010072_BAB V.pdf
Restricted to Repository staff only until 25 July 2028.

Download (322kB)
[img] Text (Daftar Pustaka)
21083010072_Daftar Pustaka.pdf

Download (213kB)
[img] Text (Lampiran)
21083010072_Lampiran.pdf
Restricted to Repository staff only until 26 July 2028.

Download (449kB)

Abstract

The Indonesian cosmetics industry recorded significant growth of 9.39% in 2023, positioning it among the fastest-growing sectors in the manufacturing domain. However, stock price movements of cosmetics companies exhibit substantial volatility, necessitating accurate volatility forecasting models to inform investment decisions and manage financial risk. This study aims to forecast stock price volatility in the Indonesian cosmetics sector using Multivariate GARCH models, specifically BEKK-GARCH and DCC-GARCH frameworks. The research utilizes daily closing prices of PT Kino Indonesia Tbk (KINO) and PT Mustika Ratu Tbk (MRAT) from January 2019 to December 2024, totaling 1,469 observations. Analysis was conducted through multivariate normality testing via Henze-Zirkler and Doornik-Hansen tests, stationarity testing via ADF and KPSS tests, white noise testing using Ljung-Box test, and heteroskedasticity testing through ARCH LM Test. Both models were estimated using Maximum Likelihood Estimation (MLE) and evaluated based on log-likelihood, AIC, BIC, and forecasting metrics including RMSE, MSE, and MAPE. Results indicate that both stocks exhibit stationary log returns with significant ARCH effects. BEKK-GARCH model demonstrates superior performance with log-likelihood of 7060.2463, AIC -14098.4926, and BIC -14042.8368, outperforming DCC-GARCH. BEKK-GARCH excels in KINO volatility prediction, while DCC-GARCH performs better for MRAT. The observed low dynamic correlation between the two stocks highlights potential for portfolio diversification. This research presents a practical framework for volatility prediction in emerging markets and develops a web-based application using Streamlit to facilitate interactive model implementation. Keywords: Multivariate GARCH, Volatility, Cosmetics Industry, BEKK-GARCH, DCC-GARCH

Item Type: Thesis (Undergraduate)
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorSaputra, Wahyu Syaifullah JauharisNIDN198608252021211003wahyu.s.j.saputra.if@upnjatim.ac.id
Thesis advisorHindrayani, Kartika MaulidaNIDN0009099205kartika.maulida.ds@upnjatim.ac.id
Subjects: H Social Sciences > HA Statistics
Q Science > QA Mathematics
Q Science > QA Mathematics > QA76.6 Computer Programming
Divisions: Faculty of Computer Science > Departemen of Data Science
Depositing User: RENALDY AL IKHSAN
Date Deposited: 28 Jul 2025 07:30
Last Modified: 29 Jul 2025 04:02
URI: https://repository.upnjatim.ac.id/id/eprint/40859

Actions (login required)

View Item View Item