Statalist


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

st: RE: RE: RE: RE: RE: outreg2 + excel + ***


From   "Lachenbruch, Peter" <[email protected]>
To   <[email protected]>
Subject   st: RE: RE: RE: RE: RE: outreg2 + excel + ***
Date   Mon, 6 Oct 2008 10:46:09 -0700

I will take this opportunity to rant a bit.

If one has specified the significance level a priori and done only the
single test, a p-value of 0.05 means the usual:  a result this extreme
or more so will occur 5% of the time.

However, most studies (rightfully) explore other variables and conduct
many tests of hypotheses.  These are suggested by the data and the
conventional meaning of significance no longer holds.  

In my analyses, I will do such tests, but I expect a much lower
(p<<0.05) level before I get excited.  It's important to recognize that
these analyses are exploratory.

Some statisticians will disagree and state that multiple testing is not
an issue.

Tony

Peter A. Lachenbruch
Department of Public Health
Oregon State University
Corvallis, OR 97330
Phone: 541-737-3832
FAX: 541-737-4001


-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Nick Cox
Sent: Monday, October 06, 2008 5:06 AM
To: [email protected]
Subject: st: RE: RE: RE: RE: outreg2 + excel + ***

Note the rampant grade inflation then! Even in the environmental
sciences, fields known for ornery systems and oddball scientists, P ~
0.1 would not be taken as anything other than a weak indication. 

Nick 
[email protected] 

Martin Weiss

Omit the -star- option in the -esttab- call and you do get this standard
behavior as in the Stata command - estimates table-. But my experience
in
the economics profession shows me that one star for ten percent is now
quite
common...

Nick Cox

Note the grade inflation: hereabouts * implies a P-value at most 0.05. 

P.S. -outreg2- is a user-written program on SSC. 

Martin Weiss

*************
cap ssc inst estout

sysuse auto, clear
eststo: reg pr w re tr, robust

esttab using myfile.csv,  star(* 0.10 ** 0.05 *** 0.01) csv replace
**************

Bastian Steingros

I have  installed the updates for *outreg2*. 

I have a regression model like:

reg dep_var ind_var1 ind_var2, robust

After entering this command in stata, I obtained the standard stata
regression table (I guess you what I mean)

Now, I want to put my results into MS EXCEL and I want to mark the
coefficients with the widely known signs like *, **, *** indicating
significance at the 10%, 5%, 1% level.

I guess it is possible to do so by using outreg2 but it seems that I do
not
understand how to use this command correctly. 

Is there anyone who knows what to do ?

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