Please use this identifier to cite or link to this item: https://repository.uksw.edu//handle/123456789/29052
Title: Suatu Aplikasi dari Model-Model Bertipe Threshold GARCH(1,1) pada Data Tokyo Stock Price Index
Authors: Puspitasari, Agnes Dhika
Keywords: distribusi skew-t;Metode ARWM;Threshold GARCH;volatilitas
Issue Date: 13-Jan-2023
Abstract: Studi ini mempelajari model-model volatilitas untuk return, yaitu TGARCH-X(1,1) dan TGARCH-CJ(1,1) yang merupakan perluasan dari model Threshold Generalized Autoregressive Conditional Heteroscedasticity (TGARCH) order (1,1) dengan menambahkan ukuran-ukuran Realized Volatility sebagai komponen eksogen. Pada model tersebut, error dari return diasumsikan berdistribusi Normal dan skew-t dari Lambert–Laurent. Analisis empiris didasarkan pada data riil, yaitu indeks saham Tokyo Stock Price Index (TOPIX) dari tahun 2004 sampai 2011. Parameter-parameter model diestimasi menggunakan metode Adaptive Random Walk Metropolis (ARWM) dalam algoritma Markov chain Monte Carlo (MCMC) yang diimplememtasikan dalam program Matlab. Berdasarkan pengamatan visual melalui trace plot (grafik nilai estimasi versus iterasi), hasil estimasi yang diperoleh menunjukkan bahwa metode ARWM efisien untuk mengestimasi model-model yang dipelajari. Selanjutnya, kecocokan model terhadap data dibandingkan menggunakan Akaike Information Criterion (AIC), yang menunjukkan bahwa model TGARCH-X(1,1)) berdistribusi skew-t memberikan pencocokan data terbaik.
This study studies the volatility models for returns, namely TGARCH-X(1,1) and TGARCH-CJ(1,1) which are extensions of the Threshold Generalized Autoregressive Conditional Heteroscedasticity (TGARCH) order (1,1) model by adding the Realized Volatility measure as an exogenous component. In this model, the error of returns is assumed to be Normal and skew-t from Lambert–Laurent. The empirical analysis based on real data, namely the Tokyo Stock Price Index (TOPIX) stock index from 2004 to 2011. The model parameters are estimated by using the Adaptive Random Walk Metropolis (ARWM) method in the Markov chain Monte Carlo algorithm (MCMC) implemented in the program Matlab. Based on visual observations through trace plots (graphs of estimated versus iteration values), the estimation results obtained show that the ARWM method is efficient for estimating the models studied. Furthermore, the fit of the model to the data was compared using the Akaike Information Criterion (AIC), which shows that the TGARCH-X(1,1) models with a skew-t distribution provide the best data match.
URI: https://repository.uksw.edu//handle/123456789/29052
Appears in Collections:T1 - Mathematics

Files in This Item:
File Description SizeFormat 
T1_662019016_Judul.pdf461.69 kBAdobe PDFView/Open
T1_662019016_Bab I.pdf
  Until 9999-01-01
295.37 kBAdobe PDFView/Open
T1_662019016_Bab II.pdf
  Until 9999-01-01
384.35 kBAdobe PDFView/Open
T1_662019016_Bab III.pdf
  Until 9999-01-01
154.81 kBAdobe PDFView/Open
T1_662019016_Bab IV.pdf
  Until 9999-01-01
618.91 kBAdobe PDFView/Open
T1_662019016_Bab V.pdf
  Until 9999-01-01
153.69 kBAdobe PDFView/Open
T1_662019016_Daftar Pustaka.pdf224.82 kBAdobe PDFView/Open
T1_662019016_Lampiran.pdf
  Until 9999-01-01
339.18 kBAdobe PDFView/Open
T1_662019016_Lisensi dan Embargo.pdf
  Restricted Access
535.05 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.