Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: Levin-Lin-Chu unit root test for panel data


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Levin-Lin-Chu unit root test for panel data
Date   Tue, 18 Feb 2003 12:28:52 -0000

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 
n.j.cox@durham.ac.uk 
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index