I am looking for a discussion beyond definitions and how it affects applied econometricians.
Sometimes with a trend, the series is I(1) and without its I(0) and vice versa. I don't want to call a variable I(0), just because I want it to be I(0). I really am looking to understand the science behind it.
While performing ADF and PP unit root test, what is the significance of such an inclusion of trend and intercept, intercept and none? Why is the lag-length criteria so important? And why don't we use that lag-length elsewhere?
What if, ADF and PP gives contradictory results?
Regarding KPSS, there results contradicts PP and ADF. Also, the problem with inclusion and exclusion of trend occurs. Any suggestions?
If the process is trend stationary, as suggested by KPSS, can it still have a UR? And how do we justify this?
Again, I understand that it depends from variable to variable, an insight with an example would help too.
Deven N Valecha
For econometric discussions not necessarily related to EViews.
1 post • Page 1 of 1
Who is online
Users browsing this forum: No registered users and 5 guests