0
$\begingroup$

I'm working with a unbalanced panel dataset of 18 countries with quarterly data from 1970 to 2014 of different characteristics per country (housing prices, education level, total population, GDP per Capita)

Now I want to figure out if these characteristics have unit roots in them.

My Question is If I should run a test per country per characteristic like this: adf.test(subset(HPI[which(HPI!="NA")],Country=="Germany"), k=4)

or if I should run the test only once per characteristic:

adf.test(HPI[which(HPI!="NA")], k=4)

Thanks!

$\endgroup$
0
$\begingroup$

Neither nor. Look for unit root tests specifially designed for panel, i.e. panel unit root test. Just search for that term for your favourite statistical package.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.