More on the correct use of omnibus tests for normality Geoffrey Poitras Simon Fraser University, Burnaby, B.C., Canada VSA lS6 Received 29 January 2005; received in revised form 1 July 2005; accepted 16 August 2005 Available online 24 January 2006
It worth to read this paper concerning the Jarque-Bera (JB) normality test: a comparison between the one-at-a-time test Skewness and Excess Kurtosis (directional) and JB test including the Urzua´s modification relating a more correct estimators of these parameter´s Expectations and Variances, along with power calculations regarding some Distributions. However I expected to find out a rather more elementary analysis, namely that the sum of two parameters (after reduction and squared) says nothing about the main problem: are these values conformable with H0, the normal Distribution? The *not rejected JB* allows us to state that data is normal? NOT AI ALL. . .