On further analysis:
This is repeatable with precisely the same
data. -lnskew0- goes round in a loop and makes
a big number bigger and bigger until it
becomes missing.
In particular,
5.7526180296e+307*2
gets mapped to
.z_
Stata tries its level best to interpret that as
a variable name, but that exchanges one error
for another.
In short: you triggered a bug.
As others have said, you can't get a normal
out of data like this anyway. Four spikes
will get mapped to four different spikes,
but four spikes do not a bell make.
Nick
[email protected]
Vivien W. Chen
> I've got a question about using lnskew0 to create a variable
> that transform
> skewness as zero. For some unknown reason, I can successfuly
> create for some
> variables, but some failed. I list the error message as
> below. Please help if
> you know the answer. Thanks!
>
> .tab mmwkdep
>
> how well
> youcould
> manage
> working
> depression | Freq. Percent Cum.
> ------------+-----------------------------------
> 0 14,474 70.00 70.00
> 1 | 4,429 21.42 91.42
> 2 | 1,293 6.25 97.67
> 3 | 481 2.33 100.00
> ------------+-----------------------------------
> Total | 296 100.00
>
>
> .lnskew0 mmwkdepLG=mmwkdep
> variable z_ not found
> r(111);
>
*
* 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/
