Automatic detection of discordant outliers via the Ueda’s method

The importance of identifying outliers in a data set is well known. Although various outlier detection methods have been proposed in order to enable reliable inferences regarding a data set, a simple but less known method has been proposed by Ueda (1996/2009). Since this new method, called Ueda’s method, has not been systematically analysed in previous research, a simulation study addressing its performance and robustness is presented. Although the method was derived assuming that the underlying data is normally distributed, its performance was analysed using data from various outlier-prone distributions commonly found in several research fields. The results obtained enable us to define the strengths and weaknesses of the method along with its limits of applicability. Furthermore, an unforeseen field of application of the method, which requires further studies was also identified.