Drexel dragonThe Math ForumDonate to the Math Forum



Search All of the Math Forum:

Views expressed in these public forums are not endorsed by Drexel University or The Math Forum.


Math Forum » Discussions » sci.math.* » sci.stat.math.independent

Topic: Jarque-Bera Test : A failed objective . . .
Replies: 0  

Advanced Search

Back to Topic List Back to Topic List  
Luis A. Afonso

Posts: 4,518
From: LIsbon (Portugal)
Registered: 2/16/05
Jarque-Bera Test : A failed objective . . .
Posted: Jul 29, 2012 5:09 PM
  Click to see the message monospaced in plain text Plain Text   Click to reply to this topic Reply

Jarque-Bera Test : A failed objective . . .


Currently in literature is stated that the Jarque-Bera test (JB) is one to assert data normality. More precisely: if data leads to a test statistics lower (or equal) to the critical value, i.e. falls into the *acceptance interval*, one should conclude that there is not sufficient evidence that normality doesn?t exist.
The statement is based on an illusory basis: because the critical values are evaluated from exact simulated random samples one is lead to assume that we are dealing with an ordinary Significance Test.
However a closer analysis is necessary.
The test is based on the sum of TWO parameters estimators: the Skewness and Excess Kurtosis after standardized (variance=1) and squared. Let be U and V, then the test statistics is JB= U + V. It´s evident that JB not rejected could be found for example with a very low U and an unlikely large V, so, providing from no normal data. In short: *accepted* test means only that at least one parameter is from likely normal data: to extend to both Skewness and Excess Kurtosis is hazardous.
In conclusion Jarque-Bera Test is not a normality test, even tough there is someone that had remarkably improved the parameters estimators. The error it is NOT at this point, of course.

Luis A. Afonso



Point your RSS reader here for a feed of the latest messages in this topic.

[Privacy Policy] [Terms of Use]

© Drexel University 1994-2014. All Rights Reserved.
The Math Forum is a research and educational enterprise of the Drexel University School of Education.