re: st: Optimal lag length with ADF

[Christopher Baum <kit.baum@bc.edu>]

<>

Barbara wrote

I want to conduct ADF tests on certain variables and would like to choose the optimal lag length according to popular information criteria like aic and bic. How can I do this in Stata? I have been trying in vain so far, conducting dfuller with different lags and then putting fitstat, but it didn't work out since fitstat never changed, always relying on the first ADF I conducted.


Forget about dfuller and pperron. Use the more appropriate dfgls. It gives you several criteria by which you may choose the optimal lag.

Kit

Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.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/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/