Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
Drexel University or The Math Forum.
|
|
Luis A. Afonso
Posts:
4,292
From:
LIsbon (Portugal)
Registered:
2/16/05
|
|
Re: The J-B test again (and again. . . )
Posted:
Jan 2, 2013 12:26 PM
|
|
Reporting to:
www.statistik-mathematik.uni-wuerzburg.de/fileadmin/100.
Johannes Hain (Aug. 2010)
Comparison of Common Tests for Normality (pg.59)
There is a different approach than Jarque-Bera?s to test normality through Skewness and Kurtosis: the tests are performed separately, each parameter in turn. A rectangle of no significance is obtained with significance alpha´=0.0253 and 0.0050 for each side. The 0.05 and 0.01 compound alpha are then obtained approximately (alpha= 1-sqrt(1-alpha´), Sidak). Note that the procedure demands a not checked feature namely the parameter´s tests independence.
Luis A. Afonso
|
|
|
|
