Prev: Serial port -received data format problem
Next: Calculate Memory
From: Yulian Isakov on 25 Jun 2010 14:42
I have been reading substantial amount of work on the Engle-Granger two step method. I have three variables in my study - one dependent and two explanatory. From my understanding, i perform a cointegrating regression (via ols) and then test the residuals for stationarity. Firstly, when testing for i(0),i(1) via adf test using the spaital-econometrics package, it does not seem to adjust the critical values according to the number of variables included. Is there a fix to this or are the critical values appropriate in the adf test for a multivariate system? Secondly, i am trying to build ecms. However, the ecm in the spatial econometrics package is based on the johansen test. How would I create the ecms after the Engle-Granger method? Is there anything available in the spatial econometrics package for that? If not would anyone direct me on how to create a function for that?

Thanks, Yulian
 | 
Pages: 1
Prev: Serial port -received data format problem
Next: Calculate Memory