Re: Who "invented" the null hypthesis?
- From: "\"Luis A. Afonso\"" <licas_@xxxxxxxxxxx>
- Date: Sun, 30 Apr 2006 01:12:08 EDT
Bob wrote:
*** Let be M a population (one dimensional case) parameter which I want to evaluate its location relative to M0
H0: M>M0______H1: M<=M0 (1)
H0: M<M0______H1: M>=M0 (2)
H0: M=M0______H1: M=/ M0 (3)
Never mind what all textbooks say. The important point is one ofthe points made by ME, that it makes NO DIFFERENCE in theproper execution of any of the tests above, if the Ho hypotheses are stated as M = M0:
H0: M=M0______H1: M<=M0 (1a)
H0: M=M0______H1: M>=M0 (2a)
H0: M=M0______H1: M=/ M0 (3a)
Explain to us any DIFFERENCE in the execution of a hypothesistest on (1) vs (1a); (2) vs (2a); or (3) vs (3a). And stay with the same subject heading you started. -- Bob. ***
WARNING
Readers shold not TRY to use this *Theory* for BINOMIAL (or other discontinuous Distribution) and ALL WILL BE WRONG!
Or when M0 is a discontinuity point of the DENSITY (of the parameter).
Bob continues his way: ignorant and stupid…
_____licas (Luis A. Afonso)
.
- Follow-Ups:
- Re: Who "invented" the null hypthesis?
- From: Reef Fish
- Re: Who "invented" the null hypthesis?
- Prev by Date: Re: Rejection of a Null Hypothesis Ho
- Next by Date: Re: Who "invented" the null hypthesis?
- Previous by thread: Re: Who "invented" the null hypthesis?
- Next by thread: Re: Who "invented" the null hypthesis?
- Index(es):
Relevant Pages
|
|
