st: is it possible to use nharvey...

[Kit Baum <baum@bc.edu>]

Alvaro said

I wonder whether this test is also available to test if variables x1t
and x2t in a simple time series could be tested to have common
trends. In this case, how to define the command? it only allow to
define one variable, which puzzles me.


You only specify one variable because in a panel, you have multiple  
time series stacked on top of one another. If you convert your series  
to that format it should work. Use the stack command to stack the two  
time series, tsset with ts# time (where ts# is 1 or 2), and then  
specify that to nharvey.

Now whether it makes any sense to do so is another thing. If these  
are two time series measuring quite different things (rather than,  
say, US and UK values of the same economic concept) you don't want to  
treat them as a panel: assuming that they might be nonstationary, you  
want to look for cointegration (a common stochastic trend) between  
them in the standard way.

Kit

Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


*
*   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/