Re: Humbly, attentively and waiting feed-back
- From: "\"Luis A. Afonso\"" <licas_@xxxxxxxxxxx>
- Date: Wed, 14 Sep 2005 22:43:47 EDT
Really you did not made allusion explicitly to my expertise in English but supporting what is said by
beliavsky@xxxxxxx and Abe Kohen corroborated by yours (3) Both writers were pointing out the insufficiency of your request. Stop focusing on the splinter in your neighbor's eye and take care of the log in your own.
I thought naturally that you included that *quality*. (Stop)
When it is said
***So, here is my feedback: multiple comparisons is a very well researched area. There are tons of books and articles (Jason Hsu's research comes to mind) which you should read instead of posting your inane questions. Guys like Tukey, Scheffe, and Bonferroni tackled it a long time ago. Start there, read up on ANOVA, do your OWN research you lazy _____, and stop asking for help if you don't like the way it is delivered to you. ***
It is an advise (are there a thing more unproductive than an advise?)... Well is said that * is a very well researched area, there are tons of books and so on * must be translated as you did: lazy, stop asking for help.
If this is really as elementary as you say. Answer, please to an even more elementary problem
***Problem
It is stated that if we know the variances of two normal populations the test that they have equal means is
________Z = | Xmean - Ymean| / s
________being s= sqrt (Xvar/nX + Yvar/nY)
(Two random samples having means Xmean, Y mean sample variances Xvar, Y var, sizes nX, NY respectively). Z has a normal standard distribution.
On the other hand
If the variances are not known but is assumed that are equal the test statistics is
__________T = |Xmean – Ymean| /s
______s= sqrt [(ssX + ssY) / (nX + nY - 2) * [1/nX + 1/nY]
being ssX = sum of the squared deviations to the mean. T has a Student nX+nY-2 df. (agree?).
My question: How are related the two estimates standard deviations of the difference of the population means. Specifically how they vary with the true standard deviations of the populations and sample sizes?
(It should be a smoothly and sweetly job for your very well-known and thoroughly educated mind, helped by tons and tons of text-books you certainly read). I am waiting
____________licas
.
- Prev by Date: Re: Remove the influence of one time series on another
- Next by Date: Re: Turning covariances into correlations: implications for sampling error
- Previous by thread: Re: Humbly, attentively and waiting feed-back
- Next by thread: Re: Humbly, attentively and waiting feed-back
- Index(es):
Relevant Pages
|