Kit Baum has posted new versions of -desrep- and -destest- in the
-desmat- package at SSC. -desrep- apparently produced errors under
version 8 due to its use of -invt- (replaced by -invttail- as of
verion 7?) and/or its use of -subscr- for the right-tail of string.
-substr(*string*,-x,.)- will apparently produce *string* even if
x>length(*string*) under version 7 and earlier. This (useful, IMHO)
behavior is considered a bug and produces an error under version 8.
Note that I don't have version 8 at present so I can't test these
things myself.
In addition, -desrep- and -destest- now respect -linesize-.
Previously, both assumed a linesize of 80. That would produce garbled
output if linesize were less than 80 which apparently depends on your
sizing of the results window.
The -desmat- package now works as intended under version 7 and
*should* work properly under version 8. Let me know if there are any
further problems and I'll see what I can do.
While I'm on the subject, I was rather hoping that -desmat- would
become redundant under version 8. Surely the Stata syntax could be
expanded to allow higher order interactions, and nested interactions
like in SAS and R. Perhaps version 9 (or a coming update)?
John Hendrickx
__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/
*
* 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/
