The problem I have is related to me wanting to account for both the presence
of serial correlation and for using generated regressors. "xtregar" fixes
the serial correlation just fine, but the estimated regressors are not being
accounted for, as I understand the procedure. Thus, I am trying to
bootstrap the standard errors. In the line below, indepvar1-indepvar3 are
all estimates from an earlier stage.
bs "xtregar depvar indepvar1 indepvar2 indepvar3 if compno ~= 0 & dir == 1 &
flag ~= 1, fe" "_b[indepvar1] _b[indepvar2] _b[indepvar3]", reps(50)
The output is returned as follows:
Bootstrap statistics
Variable | Reps Observed Bias Std. Err. [95% Conf. Interval]
---------+------------------------------------------------------------------
-
bs1 | 0 .5309594 . . . .
(N)
| . .
(P)
| . .
(BC)
---------+------------------------------------------------------------------
-
bs2 | 0 -.3184277 . . . .
(N)
| . .
(P)
| . .
(BC)
---------+------------------------------------------------------------------
-
bs3 | 0 -.3508248 . . . .
(N)
| . .
(P)
| . .
(BC)
----------------------------------------------------------------------------
-
N = normal, P = percentile, BC =
bias-corrected
Does this mean that the xtregar estimation procedure is incompatible with
Stata's canned bootstrapping techniques? What does it mean?
Many thanks,
Glen
____________________________________
Glen R. Waddell
Department of Economics, 509 PLC
1285 University of Oregon
Eugene, OR 97403-1285
phone: (541) 346-1259
fax: (541) 346-1243
http://darkwing.uoregon.edu/~waddell
[email protected]
*
* 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/
