PING  0.9
Statistical data handling and processing in production environment
var_missing_remove

Remove missing variables numeric or character from a given dataset.

%var_missing_remove(idsn, odsn,len=1400, ilib=WORK, olib=WORK);

Arguments

  • idsn : a dataset reference;
  • ilib : (option) name of the input library; by default: empty, i.e. WORK is used.
  • len : (option) max length of macro variable ; by default: 1400 is used ( max value available is 32767) .

Returns

  • odsn : name of the output dataset (in WORK library); it will contain the selection operated on the original dataset;
  • olib : (option) name of the output library; by default: empty, i.e. WORK is also used;

Examples

Let us consider the test dataset #22: breakdown | variable | start | end | fmt1_dummy | fmt2_dummy | fmt3_dummy | fmt4_dummy | fmt5_dummy | fmt6_dummy --------—|-------—|----—|--—|---------—|---------—|---------—|---------—|---------------------— label1 | DUMMY | 1 | 5 | 1 | 0 | 1 | 0 | | . label2 | DUMMY | 1 | 3 | 1 | 1 | 0 | 0 | | . label2 | DUMMY | 5 | 5 | 1 | 1 | 0 | 0 | | . label2 | DUMMY | 8 | 10 | 1 | 1 | 0 | 0 | | . label2 | DUMMY | 12 | 12 | 1 | 1 | 0 | 0 | | . label3 | DUMMY | 1 | 10 | 0 | 1 | 1 | 0 | | . label3 | DUMMY | 20 |HIGH | 0 | 1 | 1 | 0 | | . label4 | DUMMY | 10 | 20 | 0 | 0 | 0 | 1 | | . label5 | DUMMY | 10 | 12 | 0 | 1 | 1 | 0 | | . label5 | DUMMY | 15 | 17 | 0 | 1 | 1 | 0 | | . label5 | DUMMY | 19 | 20 | 0 | 1 | 1 | 0 | | . label5 | DUMMY | 30 |HIGH | 0 | 1 | 1 | 0 | | .

and run the following:

%_dstest22;
%var_missing_remove(_dstest22, TMP);

to create the output table TMP:

breakdown variable start end fmt1_dummy fmt2_dummy fmt3_dummy fmt4_dummy
label1 DUMMY 1 5 1 0 1 0
label2 DUMMY 1 3 1 1 0 0
label2 DUMMY 5 5 1 1 0 0
label2 DUMMY 8 10 1 1 0 0
label2 DUMMY 12 12 1 1 0 0
label3 DUMMY 1 10 0 1 1 0
label3 DUMMY 20 HIGH 0 1 1 0
label4 DUMMY 10 20 0 0 0 1
label5 DUMMY 10 12 0 1 1 0
label5 DUMMY 15 17 0 1 1 0
label5 DUMMY 19 20 0 1 1 0
label5 DUMMY 30 HIGH 0 1 1 0

Run macro %_example_ds_select for examples.

Notes

All character or numeric variables having missing values for alll observation in teh dataset will be removed.

References

See also