Finally an agreement was reached:
*** IN WHAT CONCERNS TWO CONTINUOUS PARAMETERS the
/two-tailed) test hypotheses is unable to state they
are equal, even though the test falls outside the
Rejection Region. The most we can attain is that ***
there are not sufficient evidence, given the data
(and gauged by the significance level), to reject the
Null Hypotheses (HOW MANY TIMES I WROTE THIS HERE;
ALWAYS DISAPROVED BY THE IMBECILE DUO?).

Luis Amaral Afonso [The moderator destroyer]

I will not claim as THE IMBECILE DUO. Jack Tomsky,
John Smith, that the Null Hypotheses can be proved
then ACCEPTED: My scientific education, intelligence
and professional carrier doesn’t allow to say such a
BARBARIOUS thing.
I left such thoughts on the behalf to the U S. genius
Stanford University produces every year.


Luis Amaral Afonso [The moderator destroyer]

Your example (concerning the test of proportions)
cannot, absolutely, be taken in account. It’s WRONG
__1__BECAUSE ONE CANNOT DRAW CONCLUSIONS FROM A ***
APPROXIMATE DISTRIBUTION ***: The problem is Binomial
not Normal at all.
__2__The occurrence |p1^ - p2^| = 0 is very rare even
though the samples have moderately large sizes.
THEN:
___One cannot (in general) state that *** WE ACCEPT
THE NULL HYPOTHESES *** (this is the exclusive
property of the IMBECILE DUO). The most we can say
(if so) is
___There is no sufficient evidence to reject the Null
Hypotheses, taking into account the data and gauged
by the significance level.
___The set of values to which ONE CANNOT REJECT THE
NULL HYPOTHESES, H0, being an INTERVAL and the
equality (?) relating H0 a single point IT FOLLOWS
THAT the statement: H0 is true, should be taken as an
illusory one.

Any other FANTASIES THE IMBECILE DUO wants to mislead
the Readers should be promptly forgotten: they are
originated from an error that they do not had way to
explain its soundness, and used some dirty tricks, I
uncover, to derive the impossible, that is, the Null
Hypotheses can be proved true, then acceptable.
One, at last a question remains: What kind of persons
can support a so evident lack of ETHICS?

Luis Amaral Afonso [The moderator ??? destroyer]

*** WHERE YOU DID FIND THE SEATEMENT?

Decision-making in statistical inference does not
require proof. That's why we have type I and type II
errors, standard errors for parameter estimators,
probabilities of misclassification, etc. If proof was
required in order to make a decision, there would be
no statistical inference. ***The Afonso equations
showed that under certain conditions, the confidence
interval for the parameter D is [0,0]. In that
particular case, he is claiming that the null
hypothesis is proved since D is known to be zero
without error. Jack (moderator) ***

My response

__1__Only APES need that people repeat (and repeat)
that the * called * my equation* did suppose the
parameters are CONTINUOUS, this is a presuppose
condition – THE IMBECILE DUO DOESN´T HAS (NOBODY HAS)
THE RIGHT TO INFECT MY RESULT applying it to a no
valid condition. THIS IS A COMPLETE UNETHICAL
BEHAVIOR THE IMBECILE DUO IS FERTILE NOBODY YET AMONG
THE READERS TRY TO PUT AN END. The interval [0,0] is
a full WORMSTOMSKY NONSENSE.

__2__THE IMBECILE DUO invent ghost theorems as long
as they think persuasive towards beginners leading to
support their FANTASIES: not to reject the Null
Hypotheses is a DECISION in its own merit. This means
that * THE STATE OF THE UNIVERSE * varied so little
that as long as one can know by data one are FORBIDEN
TO STATE THAT A CHANGE WAS OCCURRED.

__3__I had shown unmistakably that the statement THE
IMBECILE DUO I so found: to accept the null
hypotheses is so fantastic that even them TWIST,
TWIST and TURN in order to repeat the complete
BRUNDLE IT IS in Statistical terms.. They, a long
time ago, are completely aware.

Wormtomsky said:


*** The sample estimates p1^ and p2^ are not
continuous, but discrete. They have a positive
probability of being equal. When I plug that case
into Afonso's equations, I get the confidence
interval for D of [0,0].***
***************************
My response:
If X= bin (p1, n1), Y=bin(p2, n2)
The parameter space is DISCONTINUOUS:
___p1^ = j / n1, j= 0,1,2, …, n1
___p2^ = k/ n2, k= 0,1,2, …, n2

Conclusion
You are supporting me: in your mind you think that
you are correcting my point of view. You should
conclude that you are not allowed to treat (as you
did) the difference on Binomial Proportions as it was
a Continuous Parameters problem.

Furthermore
For that difference space are such that
_____D (j, k) = [j*n2 - k*n1] / (n1*n2)
_____ j = 0,1,2,…, n1
_____ k= 0,1,2,…, n2

The space consists in the * rectangle* of n1+1 times
n2+1 isolated points.

BUT THIS HAS NOTHONG TO DO WITH THE PROBLEM I POSTED
dealing with continuous parameters as they are the
mean, the standard deviation of normal populations:
WHERE THE CONFIDENCE INTERVALS ISSUED FROM THE TEST
STATISTICS DOESN`T DEGENERATE IN A SING;E POINT AND
CONSEQUENTTLY THE NULL HYPOTHESES CANNOT BE PROVED
TRUE, as all elementary text-books do not fail to
note.

Your statement that *there are a positive probability
of being equal* is a vicious and worthless one by two
reasons
__A__All person is aware (unnecessary o point out),
__B__It is (as usual your comments) OUT OF CONTEXT
NOT SAYING UN IMPORTANT THING: the probability to
have the difference p1^-p2^ = 0 are monotonically
decreasing to ZERO as the sample sizes grow. AND FROM
A APPROXIMATIVE TECHNIQUE (Binomial to Normal) INE
ARE UNABLE TO DEAW CONCLUSIONS ABOUT THE POINT:UNDER
DISCUSSION, if , for example the equation |p1 - p2|
is really attained OR NOT.

Wormtomsky:

You are so * opaque * that you think that the Type II
error evaluation says whatever to the problem under
discussion: the Null Hypotheses cannot be stated
true, or even accepted.

ERROR SUPPOSES Ha IS TRUE, A THING THAT YOU OR
WHATEVER PERSON IS UNABLE TO STATE by means an
HYPOTHESES TEST. (fail to reject Ha when Ha is
true).
You have to invent other thing to try to minimize
your BLUNDER.

To see HOW WOEMSTOMSKY IS WRONG in accepting the Null
Hypotheses, it´s suffice to take into account what
WIKIPEDIA says.



*** If the test statistic is outside the critical
region, the only conclusion is that there is not
enough evidence to reject the null hypothesis. This
is not the same as evidence in favor of the null
hypothesis. That we cannot obtain using these
arguments, since lack of evidence against a
hypothesis is not evidence for it. On this basis,
statistical research progresses by eliminating error,
not by finding the truth.***
It couldn´be clearer and smashing to WORMTOMSKY!!!
I invite the Readers to find similar material.

Luis Amaral Afonso [The moderator destroyer]

DO NOT BELIEVE IN WHAT Jack Tomsky say
From the WEB
__1__"Accepting the null hypothesis" is like
acquitting a defendant. It does NOT prove that the
null hypothesis is true, or that the defendant is
innocent. It means there is a reasonable doubt about
the defendant's guilt. In statistical testing, the
significance level, Type I risk, or alpha risk is the
"reasonable doubt." It is the chance of wrongly
rejecting the null hypothesis when it is true. In
acceptance sampling, it is the producer's risk, or
risk of wrongly rejecting a lot that meets
requirements.