Martin Weiss replied:
> Runs without a hitch for me, with and without -vers 8.2- at the top... Sure
> you ran the whole code from top to bottom? Why don`t you -trace- it to see
> where exactly it chokes?
After -set trace on-, I get this right at the end:
[...]
. ml model lf zibeta_lf (logitmu: prop = _I*) /lnphi (zg:left), robust
-------------------------------------------------------------------------------------------
begin ml ---
- version 6
- if _caller()<6 {
ml_5 `0'
exit
}
- gettoken cmd 0: 0, parse(" ,")
- local l = length(`"`cmd'"')
= local l = length(`"model"')
- if `"`cmd'"' == "clear" {
= if `"model"' == "clear" {
ml_clear `0'
exit
}
- if `"`cmd'"'=="count" {
= if `"model"'=="count" {
ml_c_d count `0'
exit
}
- if `"`cmd'"' == "trace" {
= if `"model"' == "trace" {
ml_c_d trace `0'
exit
}
- if `"`cmd'"'==substr("graph",1,max(2,`l')) {
= if `"model"'==substr("graph",1,max(2,5)) {
ml_graph `0'
exit
}
- if `"`cmd'"' == "init" {
= if `"model"' == "init" {
ml_init `0'
exit
}
- if `"`cmd'"' == substr("maximize",1,max(3,`l')) {
= if `"model"' == substr("maximize",1,max(3,5)) {
AllowEv ml_max `0'
exit
}
- if `"`cmd'"' == substr("display",1,max(2,`l')) | `"`cmd'"' ==
substr("mlout", 1,max(3,`l')) {
= if `"model"' == substr("display",1,max(2,5)) | `"model"' ==
substr("mlout", 1,max(3,5)) {
ml_mlout `0'
exit
}
- if `"`cmd'"' == substr("model",1,max(3,`l')) {
= if `"model"' == substr("model",1,max(3,5)) {
- local caller = _caller()
- version `caller': MLmodel `0'
= version 8.199999999999999: MLmodel lf zibeta_lf (logitmu: prop =
_I*) /lnphi (zg:left), robust
invalid syntax
exit
}
---------------------------------------------------------------------------------------------
end ml ---
r(198);
Work that one out.
Maarten Buis replied:
> I still have a version 8.2 on my computer, just for this type of
> situations (for those interested, I learned Stata on version 6).
> However, the example I sent runs without a problem on my machine.
> So I am stuck.
So am I, but I won't be losing sleep over it, as I need to get my
registration for Stata 9 back.
--
Clive Nicholas
[Please DO NOT mail me personally here, but at
<[email protected]>. Please respond to contributions I make in
a list thread here. Thanks!]
"My colleagues in the social sciences talk a great deal about
methodology. I prefer to call it style." -- Freeman J. Dyson.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
