Re: st: RE: R: re. ttest

[Richard Goldstein <richgold@ix.netcom.com>]

Panos,

the p-value is calculated under the assumption that the null hypothesis
is true; given that, one never "accepts" the null hypothesis; you can
only either reject it or fail to reject it

Rich

On 11/26/13, 12:18 PM, PAPANIKOLAOU P. wrote:
> Thanks Nick. The reverse would mean that if Pr((|T|>|t|)=0.94 then
> accept the null hypothesis for 0.10 critical level, is this right?
> Thanks again, Panos
> 
> 
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
> Sent: 26 November 2013 17:11
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: RE: R: re. ttest
> 
> No; that implies rejection if you use 0.10 as a critical level.
> 
> Nick
> njcoxstata@gmail.com
> 
> 
> On 26 November 2013 16:55, PAPANIKOLAOU P.
> 
>> Thanks very much, my good friend Carlo. So, when I get
> Pr((|T|>|t|)=0.06 this means that the null hypothesis would be accepted
> at 0.10, is this right?
> 
> Carlo Lazzaro
> 
>> From search ttesti- you'll get:
>> ----------------------------------------------------------
>> Title
>>
>>     [R] ttest -- Mean-comparison tests
>>
>>
>> Syntax
>>
>>     One-sample mean-comparison test
>>
>>         ttest varname == # [if] [in] [, level(#)]
>> -------------------------------------------------------------
>>
>> Replace 99 (or what you prefer) between brackets:
>>
>> ttesti 12 367 32 330, level(99)
>> ttesti 12 367 32 330, level(95)
>>
>> and you get it.
>>
> 
> PAPANIKOLAOU P.
> 
>> I want to run the ttesti using directly the 95%CI, or 99%CI, i.e. I
> want to define this option.
>> What is the best and easiest way to do this, plse?
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/