GDP growth usually is stationary, d(GPD) it is the same.
GDP in level usually is I(1).
As the unit root test shows that GDP growth is stationary then you could say that it is stationary at level. But, GDP itself may not stationary. Hope it helps.
The result depends on the first observations of the sample period, which are probably characterized by strong catch-up dynamics, Unit root tests for large GDP data could be misspecified unless the convergence phenomenon is captured with a deterministic time trend.
- what kind of data you use ?? Panel, or temporal. What kind of test you use. Generally if test say that your serie is stationnay then it is in level
I am using Panel data. I used panel Fischer type unit root test consist of both Panel ADF and Philip Perron test. My test says it is stationary. But I need research paper to prove it that it is stationary at level. Please suggest me any paper stating that GDP or any macroeconomic variable growth rate's stationarity is tested and considered it at level?
If you have directly applied Unit Root Test (means without any differencing) then your growth rate series is stationary at level
Kindly suggest any research paper supporting argument on growth rate series is stationary at level ?
if you used stata, you can mention "stata tests, stata réference and stata procedure to interpretate your . You find all in "help". Numerous paper use this practise
A time series is I(1) if the series is non-stationary and its first difference is stationary. A growth rate is the first difference of the log of a series. You have tested that the growth rate (first difference of the log is stationary). You must now apply your tests to the log of the level of the series. If this is I(1) you can then conclude that the series is I(1).
Note that the first difference of a stationary series is also stationary..
It simply means that the GDP growth rate is stationary in levels, no differencing is required in that case. The variable is Integrated of order zero.
how can i make the GDP to be stationaire at 1st different (means the Growth of GDP will be stationaire at level at the same time)...? i had run the Gross Domestic Regional Product of Indonesia by Province but i found that the GDP just only can be stationaire at 2nd different.
is it still possible for me to continue to check the cointegration with johansen system cointegration test and granger causality when my data is not stationaire? what should i do?
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Applied Statistics
- Survey Design
- Questionnaire