Peramalan Indonesian Crude Price (ICP) Dengan Bayesian Markov Switching GARCH

Herliyasari, Ria Retna (2020) Peramalan Indonesian Crude Price (ICP) Dengan Bayesian Markov Switching GARCH. Masters thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06211850010011-Master_Thesis.pdf]
Preview
Text
06211850010011-Master_Thesis.pdf

Download (1MB) | Preview

Abstract

Harga minyak mentah Indonesia atau Indonesian Crude Price (ICP) merupakan salah satu indikator makroekonomi sehingga merupakan faktor penting dalam penentuan APBN. Fluktuasi harga ICP sangat berpengaruh terhadap penerimaan dan belanja negara sehingga diperlukan prediksi ICP mendekati nilai realisasinya. ICP dapat diprediksi menggunakan analisis deret waktu karena besaran ICP merupakan observasi yang terurut. Data ICP memiliki pola yang fluktuatif sehingga memiliki variansi yang tidak homogen atau terjadi kasus heteroskedastisitas. Metode peramalan yang dapat menangkap heteroskedastisitas adalah Generalized Autoregressive Conditional Heteroscedasticity (GARCH). Namun, model GARCH tidak dapat mengetahui perubahan struktur dalam variansi. Markov switching GARCH dapat mengatasi kasus heteroskedastisitas sekaligus perubahan antar struktur variansi. Pada penelitian ini Markov switching GARCH diterapkan untuk memodelkan ICP dan dibandingkan dengan model GARCH tanpa switching. Berdasarkan nilai WAIC, MSGARCH memiliki kinerja yang lebih baik dibandingkan GARCH. Hasil peramalan menunjukkan bahwa pada Januari 2019 ICP harga ICP mengalami kenaikan menjadi US$ 67,21 dengan variansi sebesar 13,24.
=========================================================
Indonesian Crude Price (ICP) is one of the macroeconomic indicators so it is an important factor in determining the state budget. ICP price fluctuations greatly affect state revenues and expenditures so that ICP predictions are needed close to their realization value. ICP can be predicted using time series analysis because ICP quantities are ordered observations. ICP data has a fluctuating pattern so that it has a non-homogenous variance or a case of heteroscedasticity. Forecasting methods that can capture heteroscedasticity is Generalized Autoregressive Conditional Heteroscedasticity (GARCH). However, the GARCH model cannot find out structural changes in variance. Markov switching GARCH can overcome the case of heteroscedasticity as well as changes between variance structures. In this study Markov switching GARCH is applied to model ICP and compared to the GARCH model without switching. Based on the WAIC value, MSGARCH has better performance than GARCH. Forecasting results indicate that in January 2019 ICP prices increased to US$ 67.21 with a variance of 13.24.

Item Type: Thesis (Masters)
Uncontrolled Keywords: GARCH, Indonesian Crude Price, Markov switching, GARCH, Indonesian Crude Price, Markov switching
Subjects: Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry)
Q Science > QA Mathematics > QA279.5 Bayesian statistical decision theory.
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49101-(S2) Master Thesis
Depositing User: Herliyasari Ria Retna
Date Deposited: 26 Aug 2020 02:26
Last Modified: 18 Nov 2023 14:10
URI: http://repository.its.ac.id/id/eprint/81142

Actions (login required)

View Item View Item