Hi guys,
I really need your help!!!
I test time series data using ADF unit root test.
When I test one of my variables in levels I find that is non-stationary in all 3 cases( with constant and trend (insignificant), with constant(insignificant) and without trend and constant).
Taking the first difference I find that with the constant and trend, the null hypothesis of non-stationary is rejected, so is stationary. BUT with the constant only the null hypothesis is not rejected while without the constant and trend the null is rejected.
So what I have to do??? is my variable stationary in first difference or I have to take second difference??????
I am so confused with the ADF procedure, whether we have to include constant or not!!!
Please help me
Thanks in advance
ADF- Intercept and trend
Moderators: EViews Gareth, EViews Moderator
Re: ADF- Intercept and trend
Hi Guys, please I need your help, I have the same problem, in case my variable is stationnary only with intercept and trend, is it then stationay in first difference and should I detrend my serie to work with it for cointegration after that?
Thanks in advance.
Thanks in advance.
Return to “Econometric Discussions”
Who is online
Users browsing this forum: No registered users and 9 guests