fixed effect panel data model

[kazmi89]

dear members i have 10 cross sections and 13 years of data, and i want to estimate the factors behind inflation in these cross sections, however i assumed that the slope coefficients remain the same across the countries, however their are certain variables which remain fixed over time but effect the intercept of the inflation function across countries and it is also possible that these variables can b correlated across countries.. its a little introduction of what i am going to do. now inorder to estimate this i have chosen fixed effect model. i opened the panel data clicked on cross section effects as fixed and use LS and AR as model. the estimation window give me a constant term and due to this constant i am confused because:
i have learnt that fixed effect transformation is based upon time demeaned data and therefore it eliminates the time invariant terms, then why eviews is showing that intercept term? moreover it is also showing the cross sections specific constant terms in another window, as these terms should be eliminated thrugh fixed effect transformation then why it happens?
my next question is whether LSDV is also based up on the time demeaned data or not?