基于Gibbs抽样的我国汇率可加异常值的识别

The Identification of Additive Outliers Based on Gibbs Sampling in China's Exchange Rate Data

异常值会使统计分析误差增大,为了识别这些异常值,本文给出了基于Gibbs抽样识别可加异常值的方法,并用我国人民币对美元汇率的月度数据进行实证研究。

Outliers can increase the statistical analysis error.In order to identify these outliers,the paper suggested a method based on Gibbs sampling to recognize additive outliers,it also carried out empirical research using the monthly data of exchange rate of RMB against the U.S.dollars.