Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: Re: standard error and significance


From   "Kieran McCaul" <Kieran.McCaul@uwa.edu.au>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Re: standard error and significance
Date   Wed, 17 Jun 2009 07:59:15 +0800

..


A significance test will ascertain if an observed value is consistent
with the null hypothesis and if departure from the expected value under
the null could be attributed to chance.

So you need to specify the expected value, E(Y), of Y be under the null
hypothesis.

Given that I don't know exactly what you are doing, I suppose that the
test would then be z=(Y - E(Y))/SE(Y).  

______________________________________________
Kieran McCaul MPH PhD
WA Centre for Health & Ageing (M573)
University of Western Australia
Level 6, Ainslie House
48 Murray St
Perth 6000
Phone: (08) 9224-2701
Fax: (08) 9224 8009
email: Kieran.McCaul@uwa.edu.au
http://myprofile.cos.com/mccaul 
http://www.researcherid.com/rid/B-8751-2008
______________________________________________
If you live to be one hundred, you've got it made.
Very few people die past that age - George Burns

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of
jjc.li@utoronto.ca
Sent: Wednesday, 17 June 2009 7:35 AM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: Re: standard error and significance

Yes, I got the standard error of Rij after the regression by stata,  
and can calculat the standard error of Y by the formula. But how could  
I know if Y is siginificant by its standard error?


Quoting Martin Weiss <martin.weiss1@gmx.de>:

> <>
>
> The standard error of Rij will be accessible after the rgression as
-di
> _se[Rij]- which , according to your formula, is what you need...
>
> HTH
> Martin
> _______________________
> ----- Original Message ----- From: <jjc.li@utoronto.ca>
> To: "Statalist" <statalist@hsphsun2.harvard.edu>
> Sent: Wednesday, June 17, 2009 1:09 AM
> Subject: st: standard error and significance
>
>
>> Dear all,
>>
>> Here's a fomula,
>>
>> Y=1+Rij/(Si*Sj),
>>
>> where Rij is a coef. abtained form a regression; Si and Sj are    
>> obtained form a data set (let's say ten years annual data).
>>
>> Now, I am going to evaluated Y at the means of Si and Sj. The   
>> standard errors of Y can be calculated by SE(Y)=SE(Rij)/SiSj. So   
>> How could I determine if the result of Y is siginificant at 5%   
>> level, according to its standard error? How to do it by STATA?
>>
>> Thanks!
>>
>> Crystal
>>
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
>>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/




*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index