[Gretl-devel] Re: Cumulated values of matrix with missing values

Am 06.05.2021 08:38 schrieb Sven Schreiber:

> I'd say it's expected because nans propagate. Basically you don't know > what 3 + nan is. To get what you want, I was going to say "do > cum(misszero(m))", but that doesn't work on a matrix. Maybe it should? > (But remember the old fact that a NaN and a missing value is not always > the same thing.) True, 3 + nan remains unclear or is actually not defined. Fine for me if that's how it is handled -- just wanted to know what's expected in this case. Replacing nan values in matrix can be easily done by: <hansl> m = ok(m) ? m : VALUE  # VALUE is some numeric value </hansl>