Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: RE: bootstrap coefficient standard errors


From   Rubiana Chamarbagwala <rchamarb@indiana.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: bootstrap coefficient standard errors
Date   Mon, 25 Apr 2005 12:12:28 -0500

Thanks for your help Nick. Yes, both are up-to-date.
Nick Cox wrote:


Thanks. My experiments indicate that neither the number of predictors nor the length of their names
are problematic here, at least with Stata/SE. I can't see what the problem is. Are your Stata executable and ados both fully
up-to-date?
Nick n.j.cox@durham.ac.uk
Rubiana Chamarbagwala



The exact syntax I used was:

bootstrap "dprobit ftptw p_rate m_rate year88 year93 year99 hhchild f_primary f_middle f_high f_college m_primary m_middle m_high m_college
workmom lnpced asset age agesq urban lowcaste muslim season2 season3 season4 male p_rate_male m_rate_male year88_male year93_male year99_male
hhchild_male f_primary_male f_middle_male f_high_male f_college_male m_primary_male m_middle_male m_high_male m_college_male workmom_male
lnpced_male asset_male age_male agesq_male urban_male lowcaste_male muslim_male season2_male season3_male season4_male state1 state2 state3 state4
state5 state6 state7 state8 state9 state10 state11 state12 state13 state14 state15 state16 state17 state18 state19 state20 state21 state22 state23 state24 state25 state26
state27 state28 state30 state31" "_b[p_rate] _b[m_rate] _b[p_rate_male] _b[m_rate_male]", reps(1000) cluster(year_state)


Nick Cox wrote:



If you did in fact type
...
as part of your command then that is invalid syntax. If you didn't type that, then the same comment applies: we need to see _exactly_ what you typed.
Nick n.j.cox@durham.ac.uk
Rubiana Chamarbagwala





Nick - thanks for the information. Here is the syntax I used.
(1) bootstrap "dprobit y x1 x2 x3 x4 x5" "_b[x1] _b[x2]",
reps(1000)

cluster(x6) -- this worked.

(2) bootstrap "dprobit y x1 x2 x3 x4 x5 x7 x8 x9 x10 ...
x50" "_b[x1]

_b[x2]", reps(1000) cluster(x6) -- this returned "invalid syntax".

The variables in (2) above include dummies and interctions with dummies.

Nick Cox wrote:




-whelp limits- indicates various limits.
It is always worth including examples of the invalid syntax you used.
I doubt that the explanation of your problem lies in the direction you are looking. The error message you got would be normally be produced well before trying to apply the -regress-
command. However, without indications of what you actually typed it is difficult to help further.
Nick n.j.cox@durham.ac.uk
Rubiana Chamarbagwala






Hi - I am trying to estimate a probit model with lots of

independent


variables and estimate bootstrapped standard errors for 4

variables.


Stata keeps returning "invalid syntax" when I include all the independent variables. However, when I include only a few variables, it works. Does anyone know whether or not there is a limit to the number of variables one can include in a regression when one uses the bootstrap command?

*
* 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/




--
Rubiana Chamarbagwala
Assistant Professor
Department of Economics
Indiana University
100 S. Woodlawn
Bloomington, IN 47405
rchamarb@indiana.edu
http://mypage.iu.edu/~rchamarb


*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index