Paper Info
Title
Compared the Time Series Approach with Artificial Intelligent Method for Predication of Exchange Rate.
Abstract
This paper utilized the proposed PCABPN based on PSO model constructed a innovation model. The influential variables are selected based on the 27 original variables by Chang et al. model (2009) and screened through the analysis of the principal components. Principal component analysis (PCA) is performed on the original ten variables chosen from PSO model. Different variable compositions are formed so as to eliminate number of variables in order to extract 3 principal components. There are 4 variables with first component and 7 variables with the second component and 9 variables with the third component, then the variables in the different components are treated as the BPN input layer neurons. The predicted results were used to compare with GARCH and PSOBPN model, the result found the self-developed PSOPCABPN model has best prediction model.
Year
DOI
Venue
2015
10.1109/RVSP.2015.30
RVSP
Keywords
Field
DocType
ARIMA, ARCH, GARCH, PSO Model, Exchange Rate Forecasting
Autoregressive integrated moving average,Artificial intelligence,Autoregressive conditional heteroskedasticity,Statistics,Principal component analysis,Mathematics,Exchange rate
Conference
Citations 
PageRank 
References 
0
0.34
0
Authors
2
Name
Order
Citations
PageRank
Jui-Fang Chang101.69
Po-Yang Lin200.34