Abstract

ALONSO, Julio César  and  MONTENEGRO, Sebastián. A Monte Carlo Study to compare 8 normality tests for least-squares residuals following a first order autoregressive process. estud.gerenc. [online]. 2015, vol.31, n.136, pp.253-265. ISSN 0123-5923.  https://doi.org/10.1016/j.estger.2014.12.003.

The objective of this study is to assess the statistical power and size of 8 normality tests in presence of first-order autoregressive errors and different simple sizes. Using a Monte Carlo experiment, the following tests were compared: Cramér-von Mises, Anderson-Darling, Lilliefors, Pearson, Shapiro-Wilk, Shapiro-Francia, Jarque-Bera and D'Agostino-Pearson. Our results show 4 relevant findings: First, an asymmetrical effect of autocorrelation on the power and size of the tests. Second, the statistical size of all tests is affected by the autocorrelation. Third, none of the tests has greater power than the others. Fourth, the power of the normality test decreases as sample size decreases.

Keywords : Normality tests; Autocorrelation; Monte Carlo experiment; Statistical size; Statistical power.

        · abstract in Spanish | Portuguese     · text in Spanish     · Spanish ( pdf )