Forecasting performance of logistic star exchange rate model: the original and reparameterised versions / Venus Khim-Sen Liew, Ahmad Zubaidi Baharumshah and Sie-Hoe Lau
The Exponential Smooth Transition Autoregressive (ESTAR) model is widely adopted in exchange rate studies as its symmetrical distribution matches that of symmetrical exchange rate adjustment behaviour. In contrast, another specification of the STAR model, namely the LSTAR (logistic STAR) model is di...
Format: | Article |
---|---|
Language: | English |
Published: |
Universiti Teknologi MARA, Sarawak
2005
|
Online Access: | http://ir.uitm.edu.my/id/eprint/16769/ http://ir.uitm.edu.my/id/eprint/16769/1/AJ_VENUS%20KHIM-SEN%20LIEW%20JAS%2005.pdf |
id |
uitm-16769 |
---|---|
recordtype |
eprints |
spelling |
uitm-167692017-05-08T03:23:40Z http://ir.uitm.edu.my/id/eprint/16769/ Forecasting performance of logistic star exchange rate model: the original and reparameterised versions / Venus Khim-Sen Liew, Ahmad Zubaidi Baharumshah and Sie-Hoe Lau The Exponential Smooth Transition Autoregressive (ESTAR) model is widely adopted in exchange rate studies as its symmetrical distribution matches that of symmetrical exchange rate adjustment behaviour. In contrast, another specification of the STAR model, namely the LSTAR (logistic STAR) model is discarded by most researchers in priori in their exchange rate modeling exercises due to its undesired property of being asymmetrical. This study is the first of its kind in examining the validity of this hypothesis that the ESTAR exchange rate model is superior to the LSTAR exchange rate model on the basis of forecasting accuracy. Based on the experience of the adjustment process of two nominal exchange rates, we find that the hypothesis is merely theoretical since we fail to provide consistent empirical evidence in favour of the null hypothesis. This warns us that we need not be too pessimistic on the usage of the LSTAR model in exchange rate studies. In our effort to rekindle the usage of the LSTAR model, we further reparameterized the original version into the so-called absolute version, which has symmetrical distribution properties, in accordance with the well-known symmetrical adjustment process of exchange rates. The resulting ALSTAR model has proven to be a more promising model in the sense that it has improved significantly from its original version as well as the ESTAR model, which has thus far been deemed the most appropriate nonlinear exchange rate model. Universiti Teknologi MARA, Sarawak 2005 Article PeerReviewed text en http://ir.uitm.edu.my/id/eprint/16769/1/AJ_VENUS%20KHIM-SEN%20LIEW%20JAS%2005.pdf UNSPECIFIED (2005) Forecasting performance of logistic star exchange rate model: the original and reparameterised versions / Venus Khim-Sen Liew, Ahmad Zubaidi Baharumshah and Sie-Hoe Lau. Jurnal Akademik UiTM Sarawak. pp. 79-91. ISSN 0128-2635 |
repository_type |
Digital Repository |
institution_category |
Local University |
institution |
Universiti Teknologi MARA |
building |
UiTM Institutional Repository |
collection |
Online Access |
language |
English |
description |
The Exponential Smooth Transition Autoregressive (ESTAR) model is widely adopted in exchange rate studies as its symmetrical distribution matches that of symmetrical exchange rate adjustment behaviour. In contrast, another specification of the STAR model, namely the LSTAR (logistic STAR) model is discarded by most researchers in priori in their exchange rate modeling exercises due to its undesired property of being asymmetrical. This study is the first of its kind in examining the validity of this hypothesis that the ESTAR exchange rate model is superior to the LSTAR exchange rate model on the basis of forecasting accuracy. Based on the experience of the adjustment process of two nominal exchange rates, we find that the hypothesis is merely theoretical since we fail to provide consistent empirical evidence in favour of the null hypothesis. This warns us that we need not be too pessimistic on the usage of the LSTAR model in exchange rate studies. In our effort to rekindle the usage of the LSTAR model, we further reparameterized the original version into the so-called absolute version, which has symmetrical distribution properties, in accordance with the well-known symmetrical adjustment process of exchange rates. The resulting ALSTAR model has proven to be a more promising model in the sense that it has improved significantly from its original version as well as the ESTAR model, which has thus far been deemed the most appropriate nonlinear exchange rate model. |
format |
Article |
title |
Forecasting performance of logistic star exchange rate model: the original and reparameterised versions / Venus Khim-Sen Liew, Ahmad Zubaidi Baharumshah and Sie-Hoe Lau |
spellingShingle |
Forecasting performance of logistic star exchange rate model: the original and reparameterised versions / Venus Khim-Sen Liew, Ahmad Zubaidi Baharumshah and Sie-Hoe Lau |
title_short |
Forecasting performance of logistic star exchange rate model: the original and reparameterised versions / Venus Khim-Sen Liew, Ahmad Zubaidi Baharumshah and Sie-Hoe Lau |
title_full |
Forecasting performance of logistic star exchange rate model: the original and reparameterised versions / Venus Khim-Sen Liew, Ahmad Zubaidi Baharumshah and Sie-Hoe Lau |
title_fullStr |
Forecasting performance of logistic star exchange rate model: the original and reparameterised versions / Venus Khim-Sen Liew, Ahmad Zubaidi Baharumshah and Sie-Hoe Lau |
title_full_unstemmed |
Forecasting performance of logistic star exchange rate model: the original and reparameterised versions / Venus Khim-Sen Liew, Ahmad Zubaidi Baharumshah and Sie-Hoe Lau |
title_sort |
forecasting performance of logistic star exchange rate model: the original and reparameterised versions / venus khim-sen liew, ahmad zubaidi baharumshah and sie-hoe lau |
publisher |
Universiti Teknologi MARA, Sarawak |
publishDate |
2005 |
url |
http://ir.uitm.edu.my/id/eprint/16769/ http://ir.uitm.edu.my/id/eprint/16769/1/AJ_VENUS%20KHIM-SEN%20LIEW%20JAS%2005.pdf |
first_indexed |
2023-09-18T22:56:49Z |
last_indexed |
2023-09-18T22:56:49Z |
_version_ |
1777417909160640512 |