MathGroup Archive 2001

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Statistics`MeanDifferenceTest


The problem is, in conventional theory - I mean non-bayesian - both means,
i.e., the true means of the distributions from which the samples were
obtained, are supposed to be constants, albeit unknown. Therefore, you
cannot speak of probabilities concerning their relative positions.

Tomas Garza
Mexico City
----- Original Message -----
From: "Daniel Reeves" <dreeves at eecs.umich.edu>
To: mathgroup at smc.vnet.net
Subject: [mg27482] [mg27424] Statistics`MeanDifferenceTest


> I understand that given two samples s1 and s2, with means u1 and u2
> (u1>u2), MeanDifferenceTest[s1, s2, 0] gives the probability that the
> difference in sample means would be greater than or equal to u1-u2 given
> that the true means of the distributions from which they are sampled are
> equal.
>
> Knowing that, how can I compute the probability that the true mean of the
> distribution from wich s1 was sampled is greater than the true mean of the
> distribution from which s2 was sampled?
>
>
> Thanks so much!
> Daniel Reeves
>
> --    --    --    --    --    --    --    --    --    --    --    --
> Daniel Reeves               http://ai.eecs.umich.edu/people/dreeves/
>
> "You think you know when you learn,
>  are more sure when you can write,
>  even more when you can teach,
>  but certain when you can program."  -- Alan Perlis
>
>
>



  • Prev by Date: Re: Graphing Level Sets
  • Next by Date: Re: Editing formulas for the web
  • Previous by thread: Statistics`MeanDifferenceTest
  • Next by thread: RE: How to define a polyfunction?