Title :
ARIMA and neural network prediction of foreign exchange reserves
Author :
Chunhua Shi ; Huimin Wang ; Fancheng Yin ; Zhengliang Ru
Author_Institution :
State Key Lab. of Hydrol.-Water Resources & Hydraulic Eng., Hohai Univ., Nanjing, China
Abstract :
This paper is about ARIMA and neural network prediction of the foreign exchange reserves of China. Both of unit-root nonstationarity and nonlinearity are tested. In the conclusion, we show that the predictive accuracy of neural networks outperforms ARIMA in terms of the MSE and MADE criteria.
Keywords :
foreign exchange trading; mean square error methods; neural nets; prediction theory; ARIMA; China; MADE criteria; MSE criteria; foreign exchange reserves; mean absolute deviation; mean squared error; neural network prediction; unit-root nonstationarity; ISO standards; Linearity; ARIMA; neural networks; nonlinear; unit-root;
Conference_Titel :
Strategic Technology (IFOST), 2011 6th International Forum on
Conference_Location :
Harbin, Heilongjiang
Print_ISBN :
978-1-4577-0398-0
DOI :
10.1109/IFOST.2011.6021186