Vincent
>
> I am wondering if anyone has used the -levinlin- by
> Kit Baum and Fabian Bornhorst?
>
> The variable I investigated is x.
> I got the message as follows:
>
> >Levin-Lin test for x Deterministics chosen: constant
> >
> >Number of gaps in sample: 1
> >sample may not contain gaps
> >invalid syntax
> >r(198);
>
> However, before applying -levinlin-, I used tsreport
> to check my data set, the result is:
>
> >tsreport if x ,report0 panel
> >Number of gaps in sample: 0
>
> I am totally lost.
>
> My data contains missing data in x, I don't know
> wether it is the reason that I failed to apply
> -levinlin-?
You are not lost at all: you have your finger on the issue,
although there is a subtle difference here.
-levinlin- in effect strikes out your missing values and
then reports that there are gaps as a result.
Your own -tsreport- includes them -- as -if x- includes
missing values (missing is true because non-zero). So as far
as it is concerned, you have no gaps, just missing values.
So the two are working on different criteria, but are
consistent.
Nick
[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/
