[Gretl-users] forecast error calculation with "--rolling" and 1-step forecasts

Hi, I have a question about the case where I want to use the "fcast" command with the "--rolling" option (which others might call sequential instead of rolling, but that's not the point here) to do a sequence of 1-step-ahead forecasts. The doc says: "For static linear models standard errors ... incorporate ... parameter uncertainty". Given that a 1-step forecast is also static, it therefore should be possible to do that. However, I get an error about incompatible options when I combine "--rolling" and "--static". And the doc also says that with the --rolling option "forecasts are always dynamic if this is applicable". So does that mean that taking parameter uncertainty into account for a sequence of forecasts has to be done manually with an explicit loop? thanks, sven