Unit Root Test ADF
Posted: Mon May 30, 2011 8:19 am
Hi guys I confused regarding the trend component when testing for unit roots using the dickey-fuller test.
In eviews I started out with the general specification using 1st difference (intercept and the trend component). The t-statistic for trend component @trend is 0.327 and p-value 0.743. What do I conclude from this? I can remove the trend component from the model and conclude series is not a deterministic trend.
Also once I have a decided on the type of trend How do I conclude about unit roots?. If the t-stat is less than critical value it means non-stationary or is it when t-stat is above the critical value then it is non-stationary?.
thank you
In eviews I started out with the general specification using 1st difference (intercept and the trend component). The t-statistic for trend component @trend is 0.327 and p-value 0.743. What do I conclude from this? I can remove the trend component from the model and conclude series is not a deterministic trend.
Also once I have a decided on the type of trend How do I conclude about unit roots?. If the t-stat is less than critical value it means non-stationary or is it when t-stat is above the critical value then it is non-stationary?.
thank you