Abstract
| Language | English |
|---|---|
| Title of host publication | Handbook of global financial markets |
| Subtitle of host publication | transformations, dependence, and risk spillovers |
| Editors | Sabri Boubake, Duc Khuong Nguyen |
| Place of Publication | Singapore |
| Publisher | World Scientific Publishing |
| Chapter | 8 |
| Pages | 199-220 |
| Number of pages | 22 |
| ISBN (Electronic) | 9789813236653 |
| ISBN (Print) | 9789813236646 |
| DOIs | |
| Publication status | Published - Jul 2019 |
Fingerprint
Cite this
}
Currency spillover effects between the US dollar and some major currencies and exchange rate forecasts based on neural nets. / Allen, David E.; Kalev, Petko; Peiris, Shelton; Singh, Abhay K.
Handbook of global financial markets: transformations, dependence, and risk spillovers. ed. / Sabri Boubake; Duc Khuong Nguyen. Singapore : World Scientific Publishing, 2019. p. 199-220.Research output: Chapter in Book/Report/Conference proceeding › Chapter › Research › peer-review
TY - CHAP
T1 - Currency spillover effects between the US dollar and some major currencies and exchange rate forecasts based on neural nets
AU - Allen, David E.
AU - Kalev, Petko
AU - Peiris, Shelton
AU - Singh, Abhay K.
PY - 2019/7
Y1 - 2019/7
N2 - This chapter features an analysis of major currency exchange rate return spillover effects in relation to the US dollar, as constituted in US dollar terms. The Euro (EUR/USD), British Pound (GBP/USD), Chinese Yuan (CHY/USD), and Japanese Yen (JPY/USD) are modeled using the Diebold and Yilmaz (2009, 2012) spillover index metric. A rolling window of 200 days is used to capture spillover effects between the different currency pair relationships over the ten year sample of daily exchange rate returns data, from August 2005 to November 2016. We then use a neural network regression model to forecast exchange rate movements and evaluate the results on the basis of error metrics for a twenty per cent holdout sample forecast period. The spillover index analysis suggests that the greatest spillovers occur between the EUR/USD return series and the GBP/USD return series. The size and direction of spillovers changes across the sample period. This does not help for forecasting purposes, and the neural network regression models are relatively more successful in forecasting the CHY/USD relationship, possibly because of the managed nature of the Chinese currency.
AB - This chapter features an analysis of major currency exchange rate return spillover effects in relation to the US dollar, as constituted in US dollar terms. The Euro (EUR/USD), British Pound (GBP/USD), Chinese Yuan (CHY/USD), and Japanese Yen (JPY/USD) are modeled using the Diebold and Yilmaz (2009, 2012) spillover index metric. A rolling window of 200 days is used to capture spillover effects between the different currency pair relationships over the ten year sample of daily exchange rate returns data, from August 2005 to November 2016. We then use a neural network regression model to forecast exchange rate movements and evaluate the results on the basis of error metrics for a twenty per cent holdout sample forecast period. The spillover index analysis suggests that the greatest spillovers occur between the EUR/USD return series and the GBP/USD return series. The size and direction of spillovers changes across the sample period. This does not help for forecasting purposes, and the neural network regression models are relatively more successful in forecasting the CHY/USD relationship, possibly because of the managed nature of the Chinese currency.
U2 - 10.1142/9789813236653_0008
DO - 10.1142/9789813236653_0008
M3 - Chapter
SN - 9789813236646
SP - 199
EP - 220
BT - Handbook of global financial markets
A2 - Boubake, Sabri
A2 - Nguyen, Duc Khuong
PB - World Scientific Publishing
CY - Singapore
ER -