Outlier Detection Procedures in a Sample from a Gumbel Distribution with Unknown Scale Parameter

Abstract

Lalitha and Tripathi (2018) have suggested a test statistic for the detection of a pair of outliers in a sample from a Gumbel distribution with known scale parameter . The statistic is not found to be suitable while dealing with the case of unknown scale parameter . Thus, in this paper, the test statistic suggested by Lalitha and Tripathi (2018), is suitably modified by using the modified moment estimator of the scale parameter  for the detection of two single (upper/lower) outlying observations and one more test statistic is suggested for the detection of a pair of outlying observations. Their critical values and performance probabilities are obtained at different levels of significance by a simulation technique.

Keywords

Contaminant Observation, Extreme value theory, Gumbel distribution, moment estimator, Simulation, slippage alternative, trimmed sample.

  • License

    Creative Commons Attribution 4.0 (CC BY 4.0)

  • Language & Pages

    English, 31-34

  • Classification

    LCC Code: QA273.6