Hendri Adriaens wrote:
Hi again,
Another one that I do not understand:
forvalues x=1/2 {;
if `x'==1 {
local bt="bt_m";
local gt="gt_m";
};
if `x'==2 {
local bt="bt_v";
local gt="gt_v";
};
forvalues y=1/20 {;
if `y'==20 {;
local groep="b_lft95_"+"`gesl'";
};
if `y'!=20 {;
local groep="b_lft"+string((`y'-1)*5)+"_"+string(`y'*5)+"`gesl'";
};
replace `groep'=`groep'*`bt'/`gt';
};
Why does this complain like:
/gt_m invalid name
I guess it thinks the backslash is part of the string. Can I do something
about that? Notice that it doesn't seem to think this with the *`bt'...
--------------------------------------------------------------------------------
It looks more like `bt' is undefined. You can diagnose whether this is the
case by asking Stata to report what's in the bt local macro just before the
offending line:
display in smcl as input "local bt = `bt'"
replace `groep' = `groep' * `bt' / `gt'
There should be something after the equal sign in the diagnostic probe's
display line of code.
As an alternative, you can
assert "`bt'" != ""
replace `groep' = `groep' * `bt' / `gt'
The diagnostic probe will force an error if the local is undefined.
Joseph Coveney
P.S. What you seem to be trying to accomplish is normally done in Stata in a
completely different approach (i.e., in a "Stata-ish" programming style that
would result in much more efficient, readable and maintainable code). The
appearance of your code snippets over the past few posts looks like the
approach a SAS programmer coming to Stata would attempt. I don't know if
that's the case, but if it is, then I recommend a NetCourse from StataCorp
in Stata programming in order to break out of the SAS mindset that's causing
you so much of your troubles.
*
* 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/
