st: regarding negative lag coefficient in xtabond regressions

[Narasimhan Sowmyanarayanan <narasimhan.sowmyanarayanan@gmail.com>]

Hello All:

First, I want to acknowledge that this posting may not have anything
to do with STATA, probably more to do with a concept. I would
appreciate if someone can throw some light on the nature of
interpretation if the own lag in using xtabond is significant and
negative. I am particularly looking at firm profitability as a
dependent variable and the interpretation does not seem like one where
there is a growth in profits, rather suggests a decreasing nature of
profits over time. or it is simply a reflection of no-growth. Any
insight would be appreciated. I find that the correlation between
lagged values is positive however.

Thanks

Narasimhan
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/