Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | John Antonakis <John.Antonakis@unil.ch> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: coefficient interpretation in OLS |
Date | Fri, 17 Aug 2012 11:18:02 +0200 |
Hi Lynn: If you want formal (and simple) explanation run through Eq. 2 to 5c in:Antonakis, J., Bendahan, S., Jacquart, P., & Lalive, R. (2010). On making causal claims: A review and recommendations. The Leadership Quarterly, 21(6), 1086-1120.
HTH, John __________________________________________ Prof. John Antonakis Faculty of Business and Economics Department of Organizational Behavior University of Lausanne Internef #618 CH-1015 Lausanne-Dorigny Switzerland Tel ++41 (0)21 692-3438 Fax ++41 (0)21 692-3305 http://www.hec.unil.ch/people/jantonakis Associate Editor The Leadership Quarterly __________________________________________ On 17.08.2012 11:06, Nick Cox wrote:
This really is covered in every decent regression text (in your case, perhaps, an econometrics text). It's an inevitable consequence of any correlations between X4 and X1, X2, X3. Nick On Fri, Aug 17, 2012 at 11:25 PM, Lynn Lee <lynn09v@gmail.com> wrote:When I run simple OLS regression or pooled OLS regression, I find if I add more variables to the model, the coefficient on specific explanatory variable can vary in magnitude. For example, Y1=beta+beta1*X1+beta2*X2+beta3*X3+error term; Y2=alpha+alpha1*X1+ alpha2*X2+ alpha3*X3+ alpha4*X4+error term. The absolute value of estimates of beta1 or alpha1 can increase or sometimes decrease. I am not confident to explain this theoretically. Is it related to potential endogeneity issue?* * 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/