So i have this survey data that is not balanced and i have defined as a
panel using
. tsset id time
the problem is that there are like 5 or 6 other variables that are key
identifiers that need to be used in other parts of the analysis such as
seasons, housecodes, etc. The missingness for the numeric variables I
managed to fix using the lead and foreward time series operators.
what i cant fix is the alphanumeric housecode the lead/foreward operators
dont seem to like. I keep getting a type mismatch error when i do
replace hcode = L1.hcode if time > 1 & time < 4 /*there are 3 time periods*/
I was wondering if there is any cool way of filling in the grid for string
type variables in this kind of a setting.