Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: RE: joint significant


From   "Rajesh Tharyan" <[email protected]>
To   <[email protected]>
Subject   RE: st: RE: joint significant
Date   Tue, 16 Jan 2007 19:56:28 -0000

Hi,

Let me rephrase what Justin has said...

1. The null hypothesis is the hypothesis you are testing for possible
rejection under the assumption that it is true.

2. Therefore, you have to be concerned with wrongly rejecting something that
you assume is true.

3. The maximum error you are willing to accept is say .05 i.e 5 out of 100
(assuming 95% confidence = 5% significance that is where the .05 comes from)

4. The P value tell you what the significance level is. In your case it is
.08 This is higher than what you are willing to accept.

5. Therefore there is not evidence statistically to reject what you assumed
was true that is to repeat justins statement

Based on a confidence level of 95%, I would fail to reject the null
hypothesis that the estimated coefficients are jointly equal to zero.


Hope this helps
rajesh

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of White, Justin
Sent: 16 January 2007 18:51
To: [email protected]
Subject: RE: st: RE: joint significant

No.  You reject the null when the p-value is less than the set
confidence criterion.  You fail to reject the null when the p-value is
greater than the set confidence criterion.  

Examples:
P-value = 0.0890

Confidence criterion
0.10 = 90%
0.05 = 95%
0.01 = 99%

At a 90% confidence level, you would reject the null b/c 0.0890 < 0.10
At a 95% confidence level, you would fail to reject the null b/c 0.0890
> 0.05
At a 99% confidence level, you would fail to reject the null b/c 0.0890
> 0.01

You can establish you own confidence level by subtracting the p-value
from 1.

If the p-value is less than the confidence criterion, then you reject
the null.  If the p-value is greater than the confidence criterion, then
you fail to reject the null.


Justin White

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Joanne
Marshall
Sent: Tuesday, January 16, 2007 1:38 PM
To: [email protected]
Subject: RE: st: RE: joint significant

Hi again Justin,

also I have just noticed something. if my p value is 0.089, at 95%
level, 
(0.05 p value), shouldnt we reject the null (HO) because 0.089> 0.05. I 
thought we would fail to reject the null if our p value is smaller than
the 
95% p value (0.05)

Cheers Jo

>From: "White, Justin" <[email protected]>
>Reply-To: [email protected]
>To: <[email protected]>
>Subject: RE: st: RE: joint significant
>Date: Tue, 16 Jan 2007 13:26:05 -0500
>
>The H0 is the null hypothesis and H1 is the alternative hypothesis.
>Sometimes the notation can be different.  It can be H0 is the null and
>Ha is the alternative.  I always thought the H0 and Ha notation was
more
>intuitive than the H0 and H1.
>
>In this case, the null hypothesis (H0) is B2=B3=B4=0.
>The alternative hypothesis (H1) is the estimated coefficients (Bi) are
>not jointly equal to zero.
>
>Therefore, if you reject the null, you automatically accept the
>alternative which means the estimated coefficients are jointly
>significantly different from zero.
>
>If the fail to reject the null, then you are effectively accepting the
>null to be true and the estimated coefficients are jointly
>insignificantly different from zero.
>
>The joint test is different from the variable-specific test (T-test).
>Just because the F-test tells us that the variables are jointly
>different from zero does not imply that all of the estimated
>coefficients are different from zero independently.  You use a T-test
to
>determine the significance of an individual variable and use the F-test
>for joint tests.
>
>I hope this helps.
>
>
>
>Justin White
>
>-----Original Message-----
>From: [email protected]
>[mailto:[email protected]] On Behalf Of Joanne
>Marshall
>Sent: Tuesday, January 16, 2007 1:14 PM
>To: [email protected]
>Subject: RE: st: RE: joint significant
>
>Dear Justin,
>
>Thank you for making such a clear statment on the test and my result. I
>now
>understand much more than I did, and that was a much better explanation
>than
>my textbook gives!
>
>Regarding the null hypothesis, is that equivalent to H1? I always
>confused
>my H0 and H1.
>
>h0=null= estimated coefficients are jointly insignificantly different
>from
>zero. (b2=b3=b4=0)
>ho= estimated coefficients are not jointly insignificantly different
>from
>zero. (b2 not equal to b3... not equal to b4/0
>
>or is it the other way round?
>
>Thank you for your time and patience. I am finally getting this!
>
>Cheers Jo
>
>
>
> >From: "White, Justin" <[email protected]>
> >Reply-To: [email protected]
> >To: <[email protected]>
> >Subject: RE: st: RE: joint significant
> >Date: Tue, 16 Jan 2007 12:51:40 -0500
> >
> >The 91% confidence level comes from the p-value given in the Wald
test.
> >
> >
> >Here is how to interpret a p-value.  Let's say you have a p-value of
> >0.0890 from an F-test.  This tells us that given the data sample, we
>can
> >expect the estimated coefficients to be jointly equal to zero in 8.9
> >times out of 100.  This is known as Type-1 error.  If you are using a
> >confidence criterion of 95%, you are only willing to make a Type-1
>error
> >in 5 out of 100 times.  Therefore, a confidence level of 91.1% falls
> >outside of your confidence criteria and you would fail to reject the
> >null.  This means the estimated coefficients are jointly
>insignificantly
> >different from zero.
> >
> >If you are using a 95% confidence level, then you want a p-value that
>is
> >less than or equal to 0.05.  The smaller the p-value, the less likely
> >you are to make a Type-1 error.  You get the confidence level by
> >subtracting the p-value from one (1-0.0890 = 0.0911 = 91.1%)
> >
> >This would be the statement(s) you would make.....
> >Based on a confidence level of 95%, I would fail to reject the null
> >hypothesis that the estimated coefficients are jointly equal to zero.
> >
> >Or
> >
> >Based on a confidence level of 90%, I would reject the null
hypothesis
> >that the estimated coefficients are jointly equal to zero.
> >
> >Or
> >
> >I reject the null hypothesis that the estimated coefficients are
>jointly
> >equal to zero at a confidence level of 91.1%.
> >
> >Hope this helps.
> >
> >
> >Justin White
> >
> >
> >-----Original Message-----
> >From: [email protected]
> >[mailto:[email protected]] On Behalf Of Joanne
> >Marshall
> >Sent: Tuesday, January 16, 2007 12:31 PM
> >To: [email protected]
> >Subject: RE: st: RE: joint significant
> >
> >Thank you, Justin.
> >
> >"Assuming you have a 95% confidence level criterion, you will fail to
> >reject the null hypothesis that the estimated coefficients are
jointly
> >equal to zero with a confidence level of 91%."
> >
> >How can you tell we reject the null though hypothesis and where is
the
> >91%
> >from?
> > >        F(  1,   538) =    2.1
> > >            Prob > F =    0.0890
> >
> >the p value is 0.0890, which is bigger than 0.05 therefore we reject
>the
> >
> >null.
> >i am using 0.05 as the p value because it is 95% confidnece level
> >criterion.
> >is this correct?
> >thank you.
> >
> >Cheers Jo
> >
> > >From: "White, Justin" <[email protected]>
> > >Reply-To: [email protected]
> > >To: <[email protected]>
> > >Subject: st: RE: joint significant
> > >Date: Tue, 16 Jan 2007 12:23:19 -0500
> > >
> > >The Wald Test is a joint significance test.  It depends on how you
>set
> > >up the test if you want to determine if a specific coefficient has
>the
> > >appropriate sign.  The results you included tell us:
> > >
> > >Assuming you have a 95% confidence level criterion, you will fail
to
> > >reject the null hypothesis that the estimated coefficients are
>jointly
> > >equal to zero with a confidence level of 91%.
> > >
> > >There is no need to use an F-table.  The p-value given in the test
> >tells
> > >you the level of confidence.  As one of my professors told me
> > >"statistical tables are for luddites".
> > >
> > >
> > >Hope this helps.
> > >
> > >
> > >Justin White
> > >
> > >-----Original Message-----
> > >From: [email protected]
> > >[mailto:[email protected]] On Behalf Of Joanne
> > >Marshall
> > >Sent: Tuesday, January 16, 2007 12:15 PM
> > >To: [email protected]
> > >Subject: st: joint significant
> > >
> > >Dear Stata fellow,
> > >
> > >If the result which I have worked out is for a joint sign test
(Wald
> > >test)
> > >
> > >        F(  1,   538) =    2.1
> > >            Prob > F =    0.0890
> > >
> > >how can I tell if this is jointly significant or not? do I look at
> > >0.0790 or
> > >3.10 as F observ to compare with my F crit. Also from the stat.
>table,
> > >do I
> > >look for F crit under (1,538) at my desirable level on significance
>or
> > >others?
> > >
> > >Cheers
> > >Jo
> > >
> > >_________________________________________________________________
> > >MSN Hotmail is evolving - check out the new Windows Live Mail
> > >http://ideas.live.com
> > >
> > >*
> > >*   For searches and help try:
> > >*   http://www.stata.com/support/faqs/res/findit.html
> > >*   http://www.stata.com/support/statalist/faq
> > >*   http://www.ats.ucla.edu/stat/stata/
> > >
> > >
> > >*
> > >*   For searches and help try:
> > >*   http://www.stata.com/support/faqs/res/findit.html
> > >*   http://www.stata.com/support/statalist/faq
> > >*   http://www.ats.ucla.edu/stat/stata/
> >
> >_________________________________________________________________
> >MSN Hotmail is evolving - check out the new Windows Live Mail
> >http://ideas.live.com
> >
> >*
> >*   For searches and help try:
> >*   http://www.stata.com/support/faqs/res/findit.html
> >*   http://www.stata.com/support/statalist/faq
> >*   http://www.ats.ucla.edu/stat/stata/
> >
> >
> >*
> >*   For searches and help try:
> >*   http://www.stata.com/support/faqs/res/findit.html
> >*   http://www.stata.com/support/statalist/faq
> >*   http://www.ats.ucla.edu/stat/stata/
>
>_________________________________________________________________
>MSN Hotmail is evolving - check out the new Windows Live Mail
>http://ideas.live.com
>
>*
>*   For searches and help try:
>*   http://www.stata.com/support/faqs/res/findit.html
>*   http://www.stata.com/support/statalist/faq
>*   http://www.ats.ucla.edu/stat/stata/
>
>
>*
>*   For searches and help try:
>*   http://www.stata.com/support/faqs/res/findit.html
>*   http://www.stata.com/support/statalist/faq
>*   http://www.ats.ucla.edu/stat/stata/

_________________________________________________________________
MSN Hotmail is evolving - check out the new Windows Live Mail 
http://ideas.live.com

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index