Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: test of significant between coefficients


From   Nick Cox <[email protected]>
To   [email protected]
Subject   Re: st: test of significant between coefficients
Date   Tue, 27 Sep 2011 16:12:39 +0100

I think it makes about as much sense as any other significance test.

But what it means and how useful it is depend on the problem. For
example, two predictors may both be non-significant at conventional
levels because they are fighting for the same predictive share. The
way forward is likely to be not to quantify the problem by yet another
 significance test, but to try to identify the better predictor to
use.

Also, watch out. If what you are doing is looking at some significance
test results and wondering about doing more tests, then this divides
the world. Some deride this as data snooping and want to underline how
bad a strategy this is for scientific inference. Others regard this as
among the allowed tricks of intelligent data analysis in which you are
alive to all the things the data can tell you about their structure.
In many fields there is a sanctioned hypocrisy whereby you write up
the last model as if it flows automatically from the hypotheses you
now claim you had at the outset.

On Tue, Sep 27, 2011 at 3:35 PM, Andrea Rispoli <[email protected]> wrote:

> I am running a test of significance between two coefficients of the
> same OLS regression.
> My question is : if the two coefficients are not significant, does it
> still make sense to conduct the test? I am asking because sometimes
> while the individual coefficients are not significant the difference
> between them is significant, so I was trying to understand the meaning
> of this result.
*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index