i.country

[Giorgos Axiotidis]

Hello everyone,

I am working on Stata with Panel data. My panel dimension is 10 countries (Austria, Belgium, France, Greece, Germany, Ireland, Italy, the Netherlands, Portugal, Spain) and the time 10 years of weekly observations totalling 446 observations per country.

Except for my dependent and independent variables, I have created a dummy variable that has countries in the northern region as 0 and south as 1.

When I regress I use the command: xtreg Haircuts EONIA_RATE LTD DebtGDP CISS CLIF Dummy ln_eonia ln_allotment i.country, re cluster(country)

What I realise is that as expected Spain is omitted because of collinearity. The problem is that in the output instead of having 10 countries and Spain to be omitted I have results for only 9. 8 of them have normal coefficients and the Spain is omitted and Austria is not in the output at all. Why is Austria totally missing from my output? There are no errors or warnings in my outputs. This issue happens with all estimators that I use and regardless of cluster or not.

Is that normal?

Note: it is always Austria that does not appear in my results. In summary statistics, Austria is presented without any problem.

Thank you in advance,
GA

[Andrew Musau]

See listwise deletion. Most likely, Austria has missing values for all entries in one or more variables. Check your data.

[Clyde Schechter]

It does not make sense to both have country as the random effect in your -xtreg, re- model and include i.country as a covariate.