Skewness/Kurtosis

Skewness and Kurtosis SAS

In SAS, Skewness and Kurtosis are usually calculated using PROC MEANS. The procedures can produce both statistics in the same call. The procedure provides options for different methodologies.

Data Used

The following data was used in this example.

  data dat;
      input team $ points assists;
      datalines;
  A 10 2
  A 17 5
  A 17 6
  A 18 3
  A 15 0
  B 10 2
  B 14 5
  B 13 4
  B 29 0
  B 25 2
  C 12 1
  C 30 1
  C 34 3
  C 12 4
  C 11 7
  ;
  run;

Procedures Examination

By default, SAS PROC MEANS uses VARDEF option “DF”. The other options are “N”, “WEIGHT”, and “WDF. Note that the WEIGHT and WDF options produce no results, as weighted calculations are not supported in PROC MEANS for Skewness and Kurtosis.

The following shows the SAS documentation for the two measures.

Skewness

The SAS documentation for Skewness is provided here for convenience:

Kurtosis

The SAS documentation for Kurtosis is as follows:

VARDEF = DF

Skewness and Kurtosis are commonly calculated in SAS as follows:

  proc means data=dat SKEWNESS KURTOSIS;
  var points;
  run;

Output:

The above results correspond to the Type 2 methodology in R.

VARDEF = N

The N option produces the following results

  proc means data=dat SKEWNESS KURTOSIS vardef = N;
  var points;
  run;

Output:

The above results correspond to the Type 1 methodology in R.

Summary

SAS options provide for Type 1 and Type 2 Skewness and Kurtosis. Skewness Type 3 and Kurtosis Type 3 are not supported. Also Pearson’s Kurtosis is not supported.