Drop variables based on their observation count

[Julian Bausch]

Hi,

I am trying to drop variables with less than 2000 observations, in order to keep only frequent observed variables.
Could somebody help me out?

Best
JB

[Carlo Lazzaro]

Julian:
do you mean something along the following lines?

. set obs 3000
Number of observations (_N) was 0, now 3,000.

. g id=1 in 1/2000

. replace id=2 in 2001/3000

. bysort id: drop if _N<2000


.

[Nick Cox]

When you say "observations", I think you mean "non-missing values", as in Stata observation means row, record or case in the dataset and all variables are defined for all observations.

missings from the Stata Journal can report on the number of missing values, which is just the complement.

. webuse nlswork, clear
(National Longitudinal Survey of Young Women, 14-24 years old in 1968)

. missings report

Checking missings in all variables:
15082 observations with missing values

-------------------
          |      #
----------+--------
      age |     24
      msp |     16
  nev_mar |     16
    grade |      2
 not_smsa |      8
   c_city |      8
    south |      8
 ind_code |    341
 occ_code |    121
    union |   9296
   wks_ue |   5704
   tenure |    433
    hours |     67
 wks_work |    703
-------------------

. ret li

scalars:
                  r(N) =  28534

macros:
            r(varlist) : "age msp nev_mar grade not_smsa c_city south ind_code occ_code union wks_ue tenure.."

. missings report , min(5000)

Checking missings in all variables:
15082 observations with missing values

-----------------
        |      #
--------+--------
  union |   9296
 wks_ue |   5704
-----------------

. ret li

scalars:
                  r(N) =  28534

macros:
            r(varlist) : "union wks_ue"

.. net describe dm0085_2, from(http://www.stata-journal.com/software/sj20-4)

---------------------------------------------------------------------------------------------------------------
package dm0085_2 from http://www.stata-journal.com/software/sj20-4
---------------------------------------------------------------------------------------------------------------

TITLE
      SJ20-4 dm0085_2. Update: A set of utilities for ...

DESCRIPTION/AUTHOR(S)
      Update: A set of utilities for managing missing
        values
      by Nicholas J. Cox, Department of Geography,
           Durham University, Durham City, UK
      Support:  [email protected]
      After installation, type help missings
      DOI:  10.1177/1536867X20976342

INSTALLATION FILES                             (type net install dm0085_2)
      dm0085_2/missings.ado
      dm0085_2/missings.sthlp

ANCILLARY FILES                                (type net get dm0085_2)
      dm0085_2/missings.do