Search All of the Math Forum:

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

Topic: No-Null HST . . .
Replies: 9   Last Post: Feb 20, 2013 3:31 AM

 Messages: [ Previous | Next ]
 Luis A. Afonso Posts: 4,758 From: LIsbon (Portugal) Registered: 2/16/05
Re: No-Null HST . . .
Posted: Feb 19, 2013 8:39 AM

Therefore my Feb 19, 2013 5:34 AM Post, third in this thread should be read as (so sorry . . .):

It was shown, by two different procedures (Bootstrap and Intra-Permutations that the length difference between female and male spider was, in mean values, 2.21 mm.

__a) For alpha = 5% we have:

______________|______|_______|_______|_____
______________0______1.76____2.21_________Intra - Permut.
_____________________1.747___2.210________Boots

__b) For alpha = 1% we have:

______________|____|_________|_______|_____
______________0____1.63_____2.21____________Intra-Permut.
___________________1.600____2.210___________Boots

It was fine, I mean, that People - against - NHST get aware how wrong are: in fact it´s easy to conclude that at least one can warrant that the difference are AT LEAST 1.76mm (1.747) with 95% probability and 1.63 (1.60) with 99% . . .
By the reasons yet indicated the Intra-Permutation method because do not distorts data (contrarily to Bootstrap) is candidate to be the better (preferable) one. It is true that they show (almost) complete concordance.

Luis A. Afonso

Date Subject Author
2/17/13 Luis A. Afonso
2/18/13 Luis A. Afonso
2/19/13 Luis A. Afonso
2/19/13 Luis A. Afonso
2/19/13 Luis A. Afonso
2/19/13 Luis A. Afonso
2/19/13 Luis A. Afonso
2/19/13 Luis A. Afonso
2/19/13 Luis A. Afonso
2/20/13 Luis A. Afonso