{"title":"Tests for Gaussianity of a Stationary Time Series","authors":"Adnan Al-Smadi","country":null,"institution":"","volume":10,"journal":"International Journal of Electronics and Communication Engineering","pagesStart":1562,"pagesEnd":1567,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/2851","abstract":"One of the primary uses of higher order statistics in\r\nsignal processing has been for detecting and estimation of non-\r\nGaussian signals in Gaussian noise of unknown covariance. This is\r\nmotivated by the ability of higher order statistics to suppress additive\r\nGaussian noise. In this paper, several methods to test for non-\r\nGaussianity of a given process are presented. These methods include\r\nhistogram plot, kurtosis test, and hypothesis testing using cumulants\r\nand bispectrum of the available sequence. The hypothesis testing is\r\nperformed by constructing a statistic to test whether the bispectrum\r\nof the given signal is non-zero. A zero bispectrum is not a proof of\r\nGaussianity. Hence, other tests such as the kurtosis test should be\r\nemployed. Examples are given to demonstrate the performance of the\r\npresented methods.","references":null,"publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 10, 2007"}