Estimasi Tingkat Inflasi Nasional Menggunakan ARCH-GARCH Filter Kalman

Sianti, Radisha Fanni (2021) Estimasi Tingkat Inflasi Nasional Menggunakan ARCH-GARCH Filter Kalman. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

Text 06111740000045-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (1MB) | Request a copy

Abstract

Tingkat inflasi nasional merupakan salah satu indikator yang penting dalam menganalisis pertumubuhan perekonomian suatu negara. Tingkat inflasi yang tidak dikelola dengan baik dapat menyebabkan perekonomian suatu negara mengalami kemunduran. Pada data tingkat inflasi nasional digunakan model ARIMA (Autoregressive Integrated Moving Average) dan terdeteksi terdapat adanya heteroskedastisitas, sehingga digunakan model time series ARCH-GARCH (Autoregressive Conditional Heteroskedasticity-Generalized Conditional Heteroskedasticity). Model yang sesuai yaitu ARCH(1) dengan nilai MAPE (Mean Absolute Percentage Error) yang masih sangat besar yaitu 34,662%. Oleh karena itu, untuk mendapatkan nilai error yang lebih kecil dilakukan perbaikan error dengan menggunakan Filter Kalman. Hasil akhir menunjukkan bahwa Filter Kalman mampu memperbaiki hasil estimasi yang ditandai dengan nilai MAPE ARCH-Filter Kalman lebih kecil dibandingkan dengan model ARCH. Hasil estimasi terbaik pada data tingkat inflasi nasional adalah Filter Kalman polinomial derajat 2 dengan nilai Q=R=0,01 yang memiliki nilai MAPE terkecil yaitu 1,0035%.
======================================================================================================
The inflation rate is one of the most important indicators in analyzing the economic growth of a country. The inflation rate that is not managed properly causes a country’s economy to decline. The data on the national inflation rate used ARIMA (Autoregressive Integrated Moving Average) model and it is detected that there was heteroscedasticity, so the ARCH-GARCH (Autoregressive Conditional Heteroscedasticity-Generalized Conditional Heteroscedasticity) time series model was used. The appropiate model is ARCH(1) with a very large MAPE (Mean Absolute Percentage Error) value of 34,662%. Therefore, to get a smaller error rate, an error correction is made in the model with the Kalman Filter. The final result shows that the Kalman Filter is able to improve the estimation results which is indicated by the MAPE ARCH-Kalman Filter value is smaller than the ARCH model. The best estimation result on the national inflation rate data is the Kalman Filter polynomial degree 2 with a value of Q=R=0,01 which has the smallest MAPE value of 1,0035%.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	ARCH-GARCH, Kalman Filter, Inflation Rate, Filter Kalman, Tingkat Inflasi
Subjects:	Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) Q Science > QA Mathematics > QA402.3 Kalman filtering.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Radisha Fanni Sianti
Date Deposited:	26 Aug 2021 01:52
Last Modified:	26 Aug 2021 01:52
URI:	http://repository.its.ac.id/id/eprint/89735

Actions (login required)

View Item