Hi all,

I think it is not consistent to substitute missing values with zeros, because zeros are time-dated value. Missing values are genuine non-information, and it could be anything (except if the data have some structure).

Well, that's exactly why I said that this approach of Eviews (and other software) is somewhat arbitrary and it's not simply a case of an obviously missing feature on our side.

However, notice that we're talking of missing _residuals_ after estimation here, not of replacing the missing values of the series themselves with zeros.

cheers

sven