| |
 |
|
|
Science Forum Index » Statistics - Math Forum » A * Titans * Fight IV : Jack Tomky versus R A. Fisher...
Page 1 of 1
|
| Author |
Message |
| Luis A. Afonso... |
 Posted: Fri Jul 18, 2008 4:57 am |
|
|
|
Guest
|
The Fight´s outcome as expected
*** In relation to ay experiment we may speak of … the ´ null hypotheses ´ and should be noted that the null hypotheses I never proved or established, but possibly disproved, in the course of experimentation. Every experiment may be said to exist only in order to give the facts a chance of disapproving the null hypotheses. Ronald A. Fisher, The Design of Experiments, 8th ed. ,New York, Hafner Publishing Company Inc. 1966, 17.***
*************************************
Jack Tomsky, TWIST, TWIST and SHINKL, was not able to contradict what so clearly R. A. Fisher states. His argument that the SCIENTIST did not care of the alternative hypotheses is correct, all people knows. BUT THIS IS NOT BUT A KIDDS SCAPE to the point under discussion.
CLEARLY THE CORRECT DECISSION IS THIS *** THERE IS NO SUFFICIENT EVIDENCE FROM THE DATA TO REJECT THE NULL HYPOTHESES, H0. The significance level deals with the probability to reject a true H0. (To charge an innocent defendant).
*****************
Regarding the TWO-TAIL´s TEST (significance alpha) OF A PARAMETER (one or several dimensions) and taking in consideration that the PARAMETER´S SPACE comprises all possible values of this parameter, TO FAIL TO REJECT H0, means, simply, that the (true) parameter’s value is believed to be a point of the COMPLEMENTARY OF SUB-SPACE RELATIVE TO REJECT H0. It is a POINT, for example p=0, belonging to a no degenerate (not reduced to a point) CONFIDENCE INTERVAL, to which, considering a large number of similar experiments * sampling, CI´s * we expect that the fraction leading to FAIL TO REJECT H0, is precisely 1- alpha.
There are ONLY ONE *TITAN*.
Luis Amaral Afonso (The moderator destroyer) |
|
|
| Back to top |
|
| |
|
Page 1 of 1
All times are GMT - 5 Hours
The time now is Wed Oct 08, 2008 5:15 am
|
|