Re: niveau de significance

From: Stephane DRAY (
Date: Fri Nov 28 2003 - 19:20:26 MET

Hello Sara,
you can read the p-value from the following that you obtain in the listing:

number of random permutations: 5000 Observed: 1.123e+00
Histogram: minimum = 6.455e-01, maximum = 1.196e+00
number of simulations X<Obs: 4993 (frequency: 9.986e-01)
number of simulations X>=Obs: 7 (frequency: 1.400e-03)

You test the projected inertia (i.e. the quantity of Y (.xxta) that you can
explain by X (.@ob)).

The "number of simulations X>=Obs" give you the number of permutations
providing a better explanation that your real data. the p-value is this
number divided by the number of permutation (here, frequency: 1.400e-03)
=> very significant.If this number is equal to 0, then you can say that the
p-value is inferior to 1/(number of permutation).

At 10:16 28/11/2003, you wrote:
>----- Original Message -----
>From: <>Sara Mouro
>To: <>
>Sent: Thursday, November 27, 2003 4:20 PM
>Subject: le p
>Comment je peu savoir le p por le test Subspace Test du module Projectors?
>Je peu voir le histograme mais oú est le p? (que c'est refferet a le PDF
>pour ce module: «effect A is significant at p<5%»)
>Merci d'avance.
>Sara Mouro.

Stéphane DRAY

Département des Sciences Biologiques
Université de Montréal, C.P. 6128, succursale centre-ville
Montréal, Québec H3C 3J7, Canada

Tel : 514 343 6111 poste 1233
E-mail :


This archive was generated by hypermail 2b30 : Tue Sep 07 2004 - 13:45:26 MEST