ano_var_mask

Set (personal or household) variables to missing for a given list of countries.

%ano_var_mask(geo, iudb, var=, vartype=, where=, flag=yes, lib=WORK);

Arguments

• geo : list of (blank separated) countries ISO-codes; note that when geo=_ALL_, all countries will be processed, i.e. no condition on ISO-code will be imposed;
• iudb : input temporary UDB table;
• var : (option) list of variables to set to missing value .; if empty, nothing is done (i.e., masking is skipped);
• vartype : (option) type of variables passed, either personal (P), household (H) register (R) or (D); this is used to define the georeferencing variable as &vartype.B020; if empty (not recommended), the first letter of the first variable in var will be used in place of vartype;
• where : (option) additional WHERE clause used to further refine the conditions applied to select the values (observations) to set to missing;
• flag : (option) numeric or boolean (yes/no) value defined when the corresponding flag variable shall also be masked (flag=yes or any value) or not (flag=no); when flag=yes, the value used for the flag is -1;
• lib : (option) input library; default: lib=WORK.

Return

Update the table iudb by masking any/all of the (personal or household) variables that are passed to var.

Note

Let us consider two personal income variables to set to missing for SI in the personal dataset UDB_P:

%let list_of_vars = PY091G PY092G;
%udb_var_mask(SI, UDB_P, var=&list_of_vars, flag=-1, vartype=P, lib=WORK);

This is actually operating the following:

PROC SQL;
    UPDATE WORK.UDB_P
    SET     
        PY091G=., PY091G_F=-1,
        PY092G=., PY092G_F=-1
     WHERE PB020 in ("SI"); 
quit;

See also

%ano_var_round.