Analisis Risiko Sistemik Saham Subsektor Batubara Dengan Pendekatan Conditional Value-at-Risk Berbasis AR-GARCH, Regresi Kuantil, dan Stochastic Search Variable Selection

Santana, Doria Vika (2026) Analisis Risiko Sistemik Saham Subsektor Batubara Dengan Pendekatan Conditional Value-at-Risk Berbasis AR-GARCH, Regresi Kuantil, dan Stochastic Search Variable Selection. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003221067-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (10MB) | Request a copy

Abstract

Volatilitas harga saham subsektor batubara yang tinggi serta keterkaitan antar emiten berpotensi menimbulkan risiko pasar yang tidak hanya bersifat individual, tetapi juga sistemik, sehingga diperlukan pengukuran risiko yang mampu menangkap mekanisme penularan risiko pada kondisi ekstrem. Penelitian ini menganalisis risiko individual dan risiko sistemik saham subsektor batubara di Bursa Efek Indonesia menggunakan data harga penutupan harian 10 emiten periode 1 Juli 2023–30 September 2025 yang ditransformasi menjadi return. Risiko individual diestimasi melalui Value-at-Risk (VaR) pada tingkat kuantil 5% dan 1% menggunakan model AR–GARCH beserta variasinya, sedangkan risiko sistemik diukur melalui Conditional Value-at-Risk (CoVaR) dan Delta CoVaR yang diestimasi dengan regresi kuantil berbasis VaR saham lain sebagai prediktor. Selain itu, CoVaR juga diestimasi menggunakan pendekatan Stochastic Search Variable Selection (SSVS) Quantile regression untuk memperoleh model yang lebih ringkas melalui seleksi variabel probabilistik. Kinerja estimasi dievaluasi menggunakan Expected Shortfall dan uji Kupiec untuk menilai kalibrasi frekuensi pelanggaran. Hasil menunjukkan karakteristik return yang heterogen, adanya volatility clustering, dan distribusi fat-tailed. VaR 1% secara konsisten lebih dalam dibanding VaR 5% dengan CUAN sebagai saham berisiko paling tinggi, serta validitas VaR umumnya baik kecuali DSSA yang menunjukkan pelanggaran berlebih. CoVaR terkalibrasi dengan baik dan memperlihatkan peningkatan kedalaman risiko pada kondisi pasar yang lebih ekstrem, dengan sumber tekanan dominan terutama CUAN, ADRO, DSSA, dan HRUM sedangkan Delta CoVaR menegaskan kontribusi sistemik terbesar pada kondisi ekstrem berasal dari ADRO, DSSA, dan CUAN. Pendekatan CoVaR SSVS mampu mengidentifikasi keterkaitan yang paling relevan, seperti peran HRUM dan ADRO sebagai prediktor yang sering terpilih, namun performanya tidak seragam lintas saham dan kuantil. Secara keseluruhan, penggabungan VaR, CoVaR, dan Delta CoVaR memberikan kerangka yang lebih komprehensif untuk pemantauan risiko subsektor batubara, dengan implikasi praktis berupa perlunya perhatian lebih pada emiten yang berperan sebagai sumber transmisi risiko dalam kondisi pasar ekstrem.
===============================================================================================================================
High volatility in coal sub-sector stock prices and interdependencies among issuers can generate market risk that is not only idiosyncratic but also systemic; therefore, risk measures that can capture contagion under extreme conditions are required. This study analyzes individual and systemic risk in coal sub-sector stocks listed on the Indonesia Stock Exchange using daily closing prices of 10 issuers over the period 1 July 2023–30 September 2025, which are transformed into return. Individual risk is estimated using Value-at-Risk (VaR) at the 5% and 1% quantile levels based on AR–GARCH models and their variants, while systemic risk is measured using Conditional Value-at-Risk (CoVaR) and Delta CoVaR estimated via Quantile regression, where the VaR of other stocks serves as predictors. In addition, CoVaR is also estimated using Stochastic Search Variable Selection (SSVS) Quantile regression to obtain a more parsimonious model through probabilistic variable selection. Model performance is evaluated using Expected Shortfall and the Kupiec Test to assess the calibration of violation frequencies. The results indicate heterogeneous return characteristics, volatility clustering, and fat-tailed distributions. The 1% VaR is consistently more severe than the 5% VaR, with CUAN identified as the riskiest stock, and VaR validity is generally satisfactory except for DSSA, which exhibits excessive violations. CoVaR is well calibrated and shows deeper risk under more extreme market conditions, with dominant sources of stress particularly CUAN, ADRO, DSSA, and HRUM, whereas Delta CoVaR confirms that the largest systemic contributions under extreme conditions stem from ADRO, DSSA, and CUAN. The CoVaR SSVS approach is able to identify the most relevant interlinkages, such as the frequent selection of HRUM and ADRO as key predictors yet its performance is not uniform across stocks and quantile levels. Overall, combining VaR, CoVaR, and Delta CoVaR provides a more comprehensive framework for monitoring risk in the coal sub-sector, with the practical implication that greater attention should be given to issuers that act as major transmitters of risk during extreme market episodes.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	AR-GARCH, CoVaR, Risiko Sistemik, SSVS, Subsektor Batubara, AR-GARCH, Coal Subsector, CoVaR, SSVS, Systemic Risk
Subjects:	H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA30.3 Time-series analysis H Social Sciences > HA Statistics > HA31.7 Estimation H Social Sciences > HG Finance H Social Sciences > HG Finance > HG4529 Investment analysis H Social Sciences > HG Finance > HG4915 Stocks--Prices Q Science > QA Mathematics > QA274.2 Stochastic analysis Q Science > QA Mathematics > QA280 Box-Jenkins forecasting Q Science > QA Mathematics > QA402 System analysis.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Doria Vika Santana
Date Deposited:	30 Jan 2026 02:38
Last Modified:	30 Jan 2026 02:38
URI:	http://repository.its.ac.id/id/eprint/131184

Actions (login required)

View Item