------------------------------
Date: Mon, 1 Oct 2007 10:44:50 +0100 (BST)
From: Maarten buis <[email protected]>
Subject: Re: st: bootstrap error: insufficient observations ...
- --- Guido Heineck <[email protected]> wrote:
I run a multinomial logit model and want to examine differences in
predicted probabilities for different hypothetical groups of my
sample. Using the auto-data, the following code shows what I do.
<snip>
I tried several things, but so far I was not successful. I would thus
appreciate any helpful hint.
Try the following example:
*------------ example -------------------------
set more off
sysuse auto, clear
xtile lhs = price, nq(4)
gen forXmpg = foreign * mpg
capture program drop bst
program bst, rclass
mlogit lhs foreign mpg forXmpg
preserve
drop _all
set obs 4
gen foreign = _n <= 2
gen mpg = cond(mod(_n,2),11,22)
gen forXmpg = foreign * mpg
predict pr*, pr
return scalar diff11 = pr1[1]-pr1[3]
return scalar diff13 = pr3[1]-pr3[3]
return scalar diff21 = pr1[2]-pr1[4]
return scalar diff23 = pr3[2]-pr3[4]
restore
end
tempfile res
bootstrap diff11=r(diff11) diff13=r(diff13) /*
*/ diff21=r(diff21) diff23=r(diff23), /*
*/ reps(100) saving(`res') seed(999): bst
use `res', replace
*------------ end example -------------------------
(For more on how to use examples I sent to the Statalist, see
http://home.fsw.vu.nl/m.buis/stata/exampleFAQ.html )
Hope this helps,
Maarten
- -----------------------------------------
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
http://home.fsw.vu.nl/m.buis/
- -----------------------------------------