st: Re: one-tail test

[Christopher F Baum <baum@bc.edu>]

On Oct 11, 2004, at 2:33, Levy wrote:

>   How can i perform a one sided test after a linear regression?
>
>   after this regression
>         regress y x1 x2
>   i want to test if the coefficient of x2 is greater than 0
Presuming that the test statistic is of the positive persuasion, divide 
the reported p-value in half. E.g.

    headroom |   -659.463   484.5101    -1.36   0.178    -1625.788    
306.8619

This test would be insig. at 10% level or better as a two-tailed test, 
but

. di ttail(74,1.36)
.08897949

indicates that a t of this (absolute) magnitude would allow us to test 
that the coeff is negative with a significance level of 91+%, that is, 
signif. at ten percent level as a one-tailed test.

Kit Baum, Boston College Economics      baum@bc.edu
http://ideas.repec.org/e/pba1.html

*
*   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/