Olena Stavrunova <[email protected]> writes,
> I need to bootstrap [...] 5 statistics [...]. [...] I have no luck in
> obtaining bootstrap estimates for the number of statistics larger than 4. I
> use the following program, which works fine for 4 statistics :
>
> program define jobflo
> version 6.0
> [...]
> end
>
> However, if I add one more statistic for calculation so that the program
> looks like:
>
> program define jobflo1
> version 6.0
> [...]
> local tf`i' = `gh`i'' - 'jk`i''
> [...]
> post `1' `gh1998' `jk1998' `tf1998' `gh1999' `jk1999' `tf1998'
> end
There are two problems.
The first is with the line
> local tf`i' = `gh`i'' - 'jk`i''
^
|
should be `, not '
More important is how I found it, which was to run the program with
-set trace on-.
The second problem has not bitten Olena yet and perhaps it never will, but
better to be safe. It concerns the line
> post `1' `gh1998' `jk1998' `tf1998' `gh1999' `jk1999' `tf1998'
I recommend Olena change the line to read,
post `1' (`gh1998') (`jk1998') (`tf1998') (`gh1999') (`jk1999') (`tf1998')
Stata 7 would require that line look like I have just retyped it, but Olena is
using Stata 6 (or perhaps using Stata 7 in Stata-6 compatability mode).
`gh1998', `jk1998', `tf1998', `gh1999', `jk1999', and `tf1998' are all macros,
meaning each substitutes to the number itself. So,
> post `1' `gh1998' `jk1998' `tf1998' `gh1999' `jk1999' `tf1998'
might become
> post `1' 12 15.2 -17 3.25 9.1 11.7
Note the negative 17 in the middle. -post- will think it sees
> post `1' 12 (15.2-17) 3.25 9.1 11.7
and then complain that too few values were specified. This is avoided by
binding each number in parentheses ourselves;
post `1' (`gh1998') (`jk1998') (`tf1998') (`gh1999') (`jk1999') (`tf1998')
becomes
post `1' (12) (15.2) (-17) (3.25) (9.1) (11.7)
and -post- cannot become confused.
If Olena were using Stata 7, it would have forced her to use the safer syntax.
-- Bill
[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/
