Estimasi Value-at-Risk Dan Expected Shortfall Saham Sektor Energi Dan Teknologi Menggunakan Metode Hibrida ARIMA-GARCH-EVT Dengan Pendekatan POT

Hariyanto, Chielshiea Zevanya (2026) Estimasi Value-at-Risk Dan Expected Shortfall Saham Sektor Energi Dan Teknologi Menggunakan Metode Hibrida ARIMA-GARCH-EVT Dengan Pendekatan POT. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5006221024-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Pasar modal domestik rentan terhadap ketidakpastian yang memicu fluktuasi harga dan risiko ekstrem, khususnya pada sektor penyokong ekonomi seperti sektor energi dan teknologi. Penelitian ini bertujuan mengestimasi risiko ekstrem Value-at-Risk (VaR) dan Expected Shortfall (ES) pada saham sektor energi (ADRO, ITMG) dan teknologi (MTDL, EMTK) di Bursa Efek Indonesia menggunakan metode Hibrida ARIMA-GARCH-EVT dengan pendekatan Peaks Over Threshold (POT). Data penelitian menggunakan log return harga penutupan harian periode 1 Januari 2021 hingga 31 Desember 2025. Hasil analisis volatilitas membuktikan bahwa model asimetris (EGARCH dan GJR-GARCH) gagal memenuhi uji diagnostik ARCH-LM dan tidak mendeteksi adanya leverage effect berdasarkan uji Sign Bias. Oleh karena itu, model GARCH(1,1) standar dengan asumsi distribusi Student-t ditetapkan sebagai model volatilitas yang paling efisien. Sisa fluktuasi ekstrem (heavy tails) pada residual terstandarisasi kuadrat kemudian diisolasi dan dimodelkan menggunakan EVT-POT dengan asumsi Generalized Pareto Distribution (GPD) melalui penentuan threshold optimal menggunakan visualisasi Mean Residual Life Plot (MRLP). Hasil backtesting menggunakan uji Kupiec menunjukkan bahwa metode Hibrida secara konsisten memberikan estimasi VaR yang akurat dan valid pada tingkat kepercayaan 95% maupun 99% di seluruh emiten. Sebaliknya, model GARCH standar gagal memvalidasi risiko karena cenderung mengalami overestimation, terutama pada saham EMTK di tingkat kepercayaan 99%. Berdasarkan perbandingan sektoral pada kondisi ekstrem (99%), sektor teknologi mencatatkan disparitas risiko yang sangat tinggi antar emitennya, dengan EMTK sebagai saham berisiko tertinggi secara keseluruhan. Meskipun rata-rata VaR di tingkat 99% pada kedua sektor hampir setara, sektor energi memiliki nilai Expected Shortfall (ES) yang lebih dalam. Hal ini membuktikan bahwa ketika krisis ekstrem terjadi di pasar, saham sektor energi berbasis komoditas memiliki potensi kejatuhan kerugian yang jauh lebih parah.
================================================================================================================================
The domestic capital market is vulnerable to uncertainty that triggers price fluctuations and extreme risks, particularly in key economic sectors such as energy and technology. This study aims to estimate extreme risks through Value-at-Risk (VaR) and Expected Shortfall (ES) for energy (ADRO, ITMG) and technology (MTDL, EMTK) sector stocks on the Indonesia Stock Exchange using the Hybrid ARIMA-GARCH-EVT method with the Peaks Over Threshold (POT) approach. The data utilized are daily closing price log returns from January 1, 2021, to December 31, 2025. Volatility modeling results prove that asymmetric models (EGARCH and GJR-GARCH) failed the ARCH-LM diagnostic test and detected no leverage effect based on the Sign Bias test. Consequently, the standard GARCH(1,1) with a Student-t distribution was established as the most efficient volatility model. The remaining extreme fluctuations (heavy tails) in the standardized squared residuals were isolated and modeled using EVT-POT assuming a Generalized Pareto Distribution (GPD), with the optimal threshold determined via the Mean Residual Life Plot (MRLP). Backtesting evaluation via the Kupiec test demonstrates that the Hybrid method consistently provides accurate and valid VaR estimates at both 95% and 99% confidence levels across all stocks. Conversely, the standard GARCH model failed to validate risk as it tended to overestimate, especially for EMTK at the 99% level. Based on sectoral comparison under extreme conditions (99%), the technology sector exhibits high risk disparity among its issuers, with EMTK being the highest-risk stock overall. Although the average VaR at 99% for both sectors is nearly equivalent, the energy sector has a deeper Expected Shortfall (ES). This proves that during extreme market crashes, commodity-based energy stocks have the potential for far more severe loss depths.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Extreme Value Theory (EVT), GARCH, Hibrida GARCH-EVT, Peaks Over Threshold (POT), Risiko Investasi, Extreme Value Theory (EVT), GARCH, Hybrid GARCH-EVT, Investment Risk, Peaks Over Threshold (POT)
Subjects:	Q Science Q Science > QA Mathematics Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Chielshiea Zevanya Hariyanto
Date Deposited:	16 Jul 2026 07:41
Last Modified:	16 Jul 2026 07:41
URI:	http://repository.its.ac.id/id/eprint/135209

Actions (login required)

View Item