re: st: ARIMA estimation with lagged y

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

<>
KiDeuk said

I am running some time-series models and I haven't been able to understand how exactly Stata estimates my parameters. 
 
arima y L1.y , arima(2,0,0) 
 
Stata generates estimates for all parameters. But, aren't L1.y and AR(1) the same thing?


No. You have estimated the model

y_t = 0.99 y_t-1 + e_t - 0.628 e_t-1 -0.290 e_t-2

although I expect if you reproduced your full output, it would also contain a _cons and an estimate of /sigma. 
As the estimated coefficient on the LDV is essentially unity, I would expect that differencing the series would be a good idea to
avoid nonstationarity problems.

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/