Nonparametric and Computer-Intensive Resampling Tests of Randomness and Distributional Properties of Daily Changes in Foreign Exchange Measures
对五种外汇汇率日变动进行非参数和计算机密集型重抽样检验，分析其随机性、均值方差平稳性及正态性，发现非随机性、时变方差和非正态性普遍存在。
The daily changes in five measures of the foreign exchange rate are examined for their distributional characteristics. Nonparametric and computer-intensive resampling tests are performed to observe whether these changes exhibit randomness, stationarity of means and variances, and normality. Tests are done for the U.S. dollar rates and cross exchange rates of three currencies- the British pound, the German mark and the Japanese yen. Results show that non-randomness and time-varying means (the latter for dollar rates, not cross rates) generally exist. In all cases, time-varying variance and non-normality is observed. These results are compared to existing literature.