Re: Who "invented" the null hypthesis?

Dieter Folz wrote:

Who "invented" the null hypothesis, whre does it come from,
which
concepts are the basis for it? Hope, somebody can help me out.



See http://members.aol.com/jeff570/mathword.html under H for
hypothesis testing.

The following are extracts ...

"Although the use of probability in testing hypotheses is almost as
old as the study of probability, the modern terminology was largely
created by R. A. Fisher and the team of J. Neyman and E. S. Pearson in
the 1920s and -30s. "

"Null hypothesis appears in 1935 in Fisher's The Design of
Experiments. He writes, "[W]e may speak of this hypothesis as the
'null hypothesis,' and it should be noted that the null hypothesis is
never proved or established, but is possibly disproved, in the course
of experimentation." "

"This entry was contributed by John Aldrich. See also ASYMPTOTIC
RELATIVE EFFICIENCY, CHI-SQUARE, CRITICAL REGION, NEYMAN-PEARSON
(FUNDAMENTAL) LEMMA, NUISANCE PARAMETER, P-VALUE, POWER, SIGNIFICANCE,
SIMILAR REGION, SIZE, TYPE I ERROR, STUDENT'S t-DISTRIBUTION"

You should see the whole article, and possibly follow up on the other
keywords listed in the 3rd extract.

David Jones


.

Relevant Pages

  • Re: sampling question
    ... -> something new has probability 1/of being ... coin, testing the hypothesis "this coin has two ... first flip reduce the odds that the coin has a face ... -> somewhat unusual sort of hypothesis testing). ...
    (sci.math)
  • Re: Cluster analysis for beginners
    ... Rather the probability of a set of unobserved data points ... As Jeffreys famously put it: ... the outcome falls into the critical region. ... confidence intervals and hypothesis testing. ...
    (sci.stat.math)
  • Re: a hypothesis testing problem
    ... but to predict something that never happened we need a model, and that is again the null hypothesis testing. ... A model is a probability distribution. ... No - that is not an alternative hypothesis. ... You have been born: the probability ...
    (sci.stat.math)
  • Re: sampling question
    ... >population of unknown size, ... or something new has probability 1/of being something new (since ... (although I think it's a somewhat unusual sort of hypothesis testing). ... rejecting j < N at a confidence level of .001 is not the ...
    (sci.math)
Loading