SELECTION OF THE BEST UNIVARIATE NORMALITY TEST ON THE CATEGORY OF MOMENTS USING MONTE CARLO SIMULATION
Sugiyanto Sugiyanto, Etik Zukhronah · ePrints - UNY (Yogyakarta State University) · 2014
There are three univariate normality tests on the category of moment that are Geary, D’Agostino Pearson and lagrange multiplier tests. In this research, we compared them to determine which test has the highest sensitivity in the concluding test normality of the data. For the sample size n less than or equals to 30, the three tests can not be used to check for normality, whereas for sample size between 30 and 85, Geary’s test was the best tests to check normality and for sample size more than or equals to 85, the three tests were can be used to check for normality. However, for samples that have a uniform distribution, the three tests were good to check normality when the sample size more than or equals to 130.