| 基于改进多尺度阈值去噪的人民币汇率实证分析 |
| |
| 引用本文： | 姜剑梅,李星野.基于改进多尺度阈值去噪的人民币汇率实证分析[J].财务与金融,2014(5):1-5. |
| |
| 作者姓名： | 姜剑梅 李星野 |
| |
| 作者单位： | 上海理工大学 |
| |
| 基金项目： | 上海市一流学科(系统科学)建设资助项目 |
| |
| 摘 要： | 考虑到金融时间序列中噪声的干扰，有必要在金融建模研究中做适当的去噪处理。不同于传统的小波阈值去噪，本文应用改进的多尺度阈值技术对人民币/美元汇率序列进行去噪处理，并基于不同误差分布情况下综合分析，确定ARMA（1,1）-GARCH（1,1）-T为最佳拟合模型，最终给出预测效果，证实了应用新的多尺度阈值方法去噪后的汇率模型预测精度较高。
|
| 关 键 词： | 汇率 多尺度阈值 去噪 ARMA(1 1)-GARCH(1 1)-T 预测 |
Empirical Analysis of RMB Exchange Rate Based on the Improved Multi-Scale Threshold De-Noising Technology |
| |
| JIANG Jian-mei,LI Xing-ye.Empirical Analysis of RMB Exchange Rate Based on the Improved Multi-Scale Threshold De-Noising Technology[J].Accounting and Finance,2014(5):1-5. |
| |
| Authors: | JIANG Jian-mei LI Xing-ye |
| |
| Institution: | (Business School, University of Shanghai for Science and Technology, Shanghai 200093) |
| |
| Abstract: | Considering the existence of noise in the financial time series, it is necessary to deal with the noise first in modeling. Different from the traditional wavelet threshold de-noising, this paper applies the improved multi-scale threshold de-noising technology to deal with the noise in the RMB/USD exchange rate series. Based on the comprehensive analysis of different error distribution,this paper determines ARMA（1,1）-GARCH （1,1）-T to be the best fitting model. Finally it gives the predicted results and confirms that the application of the new method with multi-scale threshold de-noising model has higher prediction accuracy. |
| |
| Keywords: | Exchange Rate Multi-Scale Threshold De-Noising ARMA(1 1)-GARCH (1 1) -T Prediction |
| 本文献已被 维普 万方数据 等数据库收录！ |
