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]

st: residual in the same specification


From   "Khieu, Hinh" <[email protected]>
To   "[email protected]" <[email protected]>
Subject   st: residual in the same specification
Date   Fri, 11 Mar 2011 18:43:15 -0600

Dear statalist members,



I have a dependent variable derived as the residual of a regression below:



(1) Y = f(control variables, called X's)  - dep and independent all at time t.



Then, I run the following regression:



(2) Residual from  (1) above = f(same control variables above + 2 new variables of interest) - this time, all dependent and independent variables are at time t+1.



Could someone tell me what could be wrong with regression (2)?



In particular,



Question 1: is the residual serving as dependent variable in (2) supposed to be orthogonal to the set of control variables X's theoretically? Note that they are at 2 different times and that in reality, the results from regression (2) still show significant coefficients of a bunch of X's.



Question 2: what if regression (2) is also at time t?



I greatly appreciate any help in explaining the problem.



Thanks,
Hinh

*
*   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