The Math Forum

Search All of the Math Forum:

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

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

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: A pertinent/impertinent question. . .
Replies: 8   Last Post: Nov 13, 2013 4:02 PM

Advanced Search

Back to Topic List Back to Topic List Jump to Tree View Jump to Tree View   Messages: [ Previous | Next ]
Luis A. Afonso

Posts: 4,758
From: LIsbon (Portugal)
Registered: 2/16/05
Re: A pertinent/impertinent question. . .
Posted: Nov 13, 2013 4:02 PM
  Click to see the message monospaced in plain text Plain Text   Click to reply to this topic Reply

The Jarque-Bera inappropriateness

We have every reason, always keeping in mind that the goal is to test normality, to invite sufficiently learned people here to review critically the way the Jarque-Bera Test was constructed. Note that a former similarly proposed test is d´Agostini´s ?omnibus?: the sum of two parameter´s squared after normalizing, then a N(0,1) value. Though successfully attained this stage, every square follows a 1-degree of freedom Chi-squared, the sum a 2-degrees and finally the quantiles could be used as critical values.
But, I wonder, to test what?
Not surely the source Distribution is normal . . . It is impossible to go further than the SUM is, or is not, similar to that a normal Distribution provides. This conclusion is in complete desagrement with the current not criticized further refinement, i.e., failed to have normal intermediate stage it come on the paper: Precise finite-sample quantiles of the Jarque-Bera adjusted Lagrange multiplier test. Diethelm Wuertz and Helmuth Katzgraber (Dec. 2009). Surprisingly it seems that the critical values obtained through random simulated samples was considered sufficiently sound to ascribe normality, or not, to whatever sample tested this way. For me is indisputable that an unsolved main problem was shift to other that is rather second rank in importance. Is people comfortable with a procedure that is unable to answer what it was proposed?
Or I am wrong?

Luis A. Afonso

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

[Privacy Policy] [Terms of Use]

© The Math Forum at NCTM 1994-2018. All Rights Reserved.