摘要
为了描述汇率变动之间的远程相关行为，本文提出了刻划浮动汇率的一种新模型并给出了具体的建模方法，实验结果说明该模型是有意义的．
As stated early in many literatures,the variability of flexible exchange rate may not be independent but exhibit the long-run dependent behavior. The classical Random-Walk model can,t characterize such behav-ior; ARCH(GARCH) and ARFIMA models can describe the dependence of time series,but they are difficult to build. In this paper,we propose a new and simple mode1 named 'Increment-Fractional Gaussian Noise (I-FGN) process', which can characterize the long-run dependence of the flexible exchange rate,furthermore,the model is easy to describe and build. In this paper,we also give the detai1ed methods for building modelswith I-FGN process. Using the true data of JPY/USD exchange rate,we bui1d an l-FGN model to show thevalidity of our model. From the experiment resuIt,we conclude that the increment of JPY/USD exchange rate exhibit long-run positive dependence.
出处
《经济数学》
1998年第Z1期17-25,共9页
Journal of Quantitative Economics
关键词
浮动汇率
随机行走模型
小波
分数高斯噪声
flexible exchange rate, random walk model, wavelet,fractional gaussian noise
