Dear all,
I am using the �ipshin� procedure to test for unit
roots in panel data (28 firms over 14 time periods)
for variables such as log of output, log of labour
hours, etc.
However, I am not very clear on how to specify lags.
For instance, testing for I(1) in the case of no
serial correlation (constant, no trend):
in order to estimate equation dy(i,t) = a(i) +
b(i)*y(i,t-1) + e(i,t) � where dy(i,t) is y(i,t) �
y(i, t-1) - shall I specify lag = 0?
Also, I am not sure whether my sample size is
sufficient for the case of serially correlated errors.
I would be very grateful if you could please help me.
Many thanks in advance.
Federica Maiorano
PhD student
Department of Economics
City University
___________________________________
Nuovo Yahoo! Messenger: E' molto pi� divertente: Audibles, Avatar, Webcam, Giochi, Rubrica� Scaricalo ora!
http://it.messenger.yahoo.it
*
* 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/
