Darmanto, Isnani Darti, Suci Astutik, Nurjannah
This study aims to model the volatility of the top three pharmaceutical sector stocks in Indonesia using a two-stage modeling approach. The first stage involves GARCH and its variations, followed by residual modeling using the GPD-based EVT approach. The three stocks analyzed are Mitra Keluarga Karyasehat Tbk (IDX: MIKA), Kalbe Farma Tbk (IDX: KLBF), and Medikaloka Hermina Tbk (IDX: HEAL). The dataset comprises return data spanning July 1, 2019, to April 30, 2025 (a total of 1,412 observations), sourced from Yahoo Finance. The findings reveal that the best-fitting models for capturing volatility are ARMA(0,1)-GARCH(1,1) for MIKA.JK, ARMA(1,2)-TGARCH(1,1) for KLBF.JK, and ARMA(0,0)-GJRGARCH(1,1) for HEAL.JK. Backtesting and the Kupiec test confirm that these models provide highly accurate VaR predictions at the 1%, 2.5%, and 5% quantiles. © 2025, Hungarian Central Statistical Office. All right resevered.
Department of Statistics, Faculty of Mathematics and Natural Sciences, Universitas Brawijaya, Jl. Veteran, Malang, Indonesia; Department of Mathematics, Faculty of Mathematics and Natural Sciences, Universitas Brawijaya, Jl. Veteran, Malang, Indonesia
