Re: st: how zandrews command chooses k lags by t-stat test

[Mark <statalist528h5q@feldmark.com>]

Thanks for the confirmation.  Actually I am working on a routine that 
incorporates Lumsdaine and Papell (1997)'s extensions to 2 breaks, and 
just for fun, lets you specify an arbitrary number of breaks.  So based 
on a command line option, it will provide statistics for the test with 
1, 2, or more breaks.

I am currently deciding how to handle some minor issues, writing a help 
page, and tracking down critical values, which are a problem to some 
extent, because they dont exist for some combinations of types of 
breaks, nor at all for > 2 breaks.  But otherwise, a test version is 
basically coded up already.  So in case you dont get around to updating 
-zandrews-, I may be able to provide an interesting alternative one of 
these days.

Mark

Lumsdaine, Robin and David H. Papell (1997) “Multiple Trend Breaks and 
the Unit-Root Hypothesis,” The Review of Economics and Statistics, Vol. 
79, No. 2 (May 1997), pp. 212-218.

Kit Baum wrote:

> Mark is quite right. As the help file for -zandrews- indicates, this 
> routine was translated from RATS code (available from 
> http://www.estima.com/procs_alpha.shtml) more or less wholesale, with 
> one or two corrections. That RATS code does not incorporate the 
> recommended determination of lag length as separate for each \lambda 
> (break point). Although the original JBES article was consulted in the 
> development of -zandrews-, the excellent point Mark raises was not 
> incorporated into the Stata code. I will put that fix on the queue of 
> desirable enhancements to my routines. Thanks to Mark for bringing it to 
> my attention.
*
*   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/