Modelling gold price using ARIMA-TGARCH
Statistical models can be used to characterize numerical data so as to understand its behavior and pattern. Gold price model, for example, can give signals to investors as to when they should enter and/or exit the market. To find an appropriate gold price model, it is crucial to choose a model that...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Hikari
2016
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/14660/ http://umpir.ump.edu.my/id/eprint/14660/ http://umpir.ump.edu.my/id/eprint/14660/ http://umpir.ump.edu.my/id/eprint/14660/1/Modelling%20Gold%20Price%20using%20ARIMA%20%E2%80%93%20TGARCH.pdf |
| Summary: | Statistical models can be used to characterize numerical data so as to understand its behavior and pattern. Gold price model, for example, can give signals to investors as to when they should enter and/or exit the market. To find an appropriate gold price model, it is crucial to choose a model that reflects the pattern of the price movement so as to make the model fit and adequate. This study examines the performances of ARIMA-TGARCH with five innovations in
modeling and forecasting gold prices. The innovations considered include Gaussian, Student’s-t, skewed Student’s-t, generalized error distribution and skewed generalized error distribution. Using daily gold price data from the years 2003 to 2014, this study concluded that a hybrid ARIMA(0,1,0)-TGARCH(1,1) with t-innovation was the best model due to the existence of leverage effect and
heavier tail characteristics in the data.
|
|---|