Bookmark and Share

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

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

Re: st: Including constant?

From   Daniel Marcelino <>
Subject   Re: st: Including constant?
Date   Fri, 3 Jun 2011 14:59:59 -0300

This discussion lead me think about the point.
In theory, the constant could be removed from models. Although, in
practice we do not usually do that. The p-value for a particular
constant may be used to determine whether keep or drop it. If you drop
the constant your model should look likes this  Y = b1 X, of course
the problem here is that Y should be interpreted as 0 when X is 0.


On Thu, Jun 2, 2011 at 12:15 PM, Richard Williams
<> wrote:
> At 12:31 AM 6/2/2011, Nils Braakmann wrote:
>> Interesting to learn about these disciplinary differences. In
>> economics the most common thing nowadays seems to be to only report
>> coefficients/results for your main variable of interest, omit results
>> for control variables, constants etc.  and just mention that these
>> were included somewhere in the text/a table footnote.
>> Cheers,
>> Nils
> I am not crazy about that but yes, especially if you have a zillion
> variables, I can see doing it that way. I would hope that the complete
> results are available upon request. One problem with such an approach is
> that it might obscure the fact that the variables of interest have pretty
> small effects compared to everything else. It might also hide suspicious
> results that could call the rest of the findings into question.
> -------------------------------------------
> Richard Williams, Notre Dame Dept of Sociology
> OFFICE: (574)631-6668, (574)631-6463
> HOME:   (574)289-5227
> EMAIL:  Richard.A.Williams.5@ND.Edu
> WWW:
> *
> *   For searches and help try:
> *
> *
> *

Daniel Marcelino
Skype: d_marcelino

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index