Regularizing BWGC/NGARCH Model by Quantum Minimization

Author

Chang, Bao Rong ; Tsai, Hsiu Fen ; Chen, Shi Huang ; Chen, Yu Chang ; Tseng, Yu-Kuo

Author_Institution

Dept. of Comput. Sci. & Inf. Eng., Nat. Taitung Univ.

Volume

2

fYear

2006

fDate

Aug. 30 2006-Sept. 1 2006

Firstpage

250

Lastpage

253

Abstract

A hybrid BPNN-weighted grey-C3LSP prediction (BWGC) is used for resolving the overshooting phenomenon significantly; however, it may lose the localization once volatility clustering occurs. Thus, we propose a compensation to deal with the time-varying variance in the residual errors, that is, incorporating a non-linear generalized autoregressive conditional heteroscedasticity (NGARCH) into BWGC, and quantum minimization is employed to regularize the smoothing coefficients for both BWGC and NGARCH to effectively improve model´s robustness as well as to highly balance the generalization and the localization

Keywords

autoregressive processes; backpropagation; economic forecasting; forecasting theory; grey systems; least squares approximations; minimisation; neural nets; stock markets; backpropagation neural net; hybrid BPNN-weighted grey-C3LSP prediction model; nonlinear generalized autoregressive conditional heteroscedasticity model; overshooting phenomenon; quantum minimization; residual error; time-varying variance; volatility clustering; Computer errors; Computer science; Machine intelligence; Neural networks; Neurons; Predictive models; Robustness; Smoothing methods; Tail; Transfer functions;

fLanguage

English

Publisher

ieee

Conference_Titel

Innovative Computing, Information and Control, 2006. ICICIC '06. First International Conference on

Conference_Location

Beijing

Print_ISBN

0-7695-2616-0

Type

conf

DOI

10.1109/ICICIC.2006.338

Filename

1691974