Cumulant plots and goodness-of-fit tests for the inverse Gaussian distribution

摘要: This paper uses a standardized version of the logarithm empirical moment generating function in order to construct plots for assessing appropriateness inverse Gaussian distribution. Variability is added by utilizing asymptotic and finite-sample results. The have linear scales do not rely on use tables or special functions. In addition, they are equivalent goodness-of-fit test whose critical values obtained from fitted equations involving sample size estimated shape parameter Three data sets used illustrate plots. A similar also proposed found through parametric bootstrap. An extensive simulation study shows that new tests maintain good stability level high power across wider range distributions sizes than other tests.