[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: dropped observations in bootstrap

From   Michael McCulloch <>
Subject   Re: st: dropped observations in bootstrap
Date   Sun, 7 Oct 2007 12:09:48 -0700

Brilliant! It makes perfect sense, that when the bootstrap is re-sampling it's attempting to generate variables that already exist. Thank you so much.

--- Maarten wrote:
 > One step you can take is to call the program -msm- you created,
 > outside -bootstrap- and see if it works, you can call it a second
 > time and you will see that it no longer works because several
 > variables are-generate-ed which already exist (where created during
 > the first call to -msm-). How did I know to call -msm- twice? I
 > made the same mistake you made many many times...
--- Michael McCulloch <> wrote:
 My temporary elation is tempered by the nagging discomfort of not
 knowing why the use of temporary variables resolved the issue. I'll
 proceed with your suggested -trace- approach. Many thanks, Michael
So, your program needs to create a number of variables. It leaves those
variables behind after it has finished. With -bootstrap- you run your
program multiple times (as many as you specify in the -reps()-
options). So now when you try to run the program multiple times, -msm-
tries to -generate- variables that already exist (were made by the
previous call to -msm-). This causes errors, as a result it returns no
results to -bootstrap- causing to display red X instead green dots.

Hope this helps,

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Want ideas for reducing your carbon footprint? Visit Yahoo! For Good
* For searches and help try:
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index