Previous Page | Next Page

The UNIVARIATE Procedure

Calculating Percentiles

The UNIVARIATE procedure automatically computes the 1st, 5th, 10th, 25th, 50th, 75th, 90th, 95th, and 99th percentiles (quantiles), as well as the minimum and maximum of each analysis variable. To compute percentiles other than these default percentiles, use the PCTLPTS= and PCTLPRE= options in the OUTPUT statement.

You can specify one of five definitions for computing the percentiles with the PCTLDEF= option. Let be the number of nonmissing values for a variable, and let represent the ordered values of the variable. Let the th percentile be , set , and let

     

where is the integer part of np, and is the fractional part of np. Then the PCTLDEF= option defines the th percentile, , as described in the following table.

PCTLDEF

Description

Formula

1

weighted average at

   

where is taken to be

2

observation numbered closest to np

3

empirical distribution function

4

weighted average aimed

 

at

where is taken to be

5

empirical distribution function with averaging

Weighted Percentiles

When you use a WEIGHT statement, the percentiles are computed differently. The 100th weighted percentile is computed from the empirical distribution function with averaging:

     

where is the weight associated with and is the sum of the weights.

Note that the PCTLDEF= option is not applicable when a WEIGHT statement is used. However, in this case, if all the weights are identical, the weighted percentiles are the same as the percentiles that would be computed without a WEIGHT statement and with PCTLDEF=5.

Confidence Limits for Percentiles

You can use the CIPCTLNORMAL option to request confidence limits for percentiles, assuming the data are normally distributed. These limits are described in Section 4.4.1 of Hahn and Meeker (1991). When , the two-sided confidence limits for the th percentile are

     

where is the sample size. When , the two-sided confidence limits for the th percentile are

     

One-sided confidence bounds are computed by replacing by in the appropriate preceding equation. The factor is related to the noncentral distribution and is described in Owen and Hua (1977) and Odeh and Owen (1980). See Example 4.10.

You can use the CIPCTLDF option to request distribution-free confidence limits for percentiles. In particular, it is not necessary to assume that the data are normally distributed. These limits are described in Section 5.2 of Hahn and Meeker (1991). The two-sided confidence limits for the th percentile are

     

where is the th order statistic when the data values are arranged in increasing order:

     

The lower rank and upper rank are integers that are symmetric (or nearly symmetric) around , where is the integer part of and is the sample size. Furthermore, and are chosen so that and are as close to as possible while satisfying the coverage probability requirement,

     

where is the cumulative binomial probability,

     

In some cases, the coverage requirement cannot be met, particularly when is small and is near 0 or 1. To relax the requirement of symmetry, you can specify CIPCTLDF(TYPE = ASYMMETRIC). This option requests symmetric limits when the coverage requirement can be met, and asymmetric limits otherwise.

If you specify CIPCTLDF(TYPE = LOWER), a one-sided lower confidence bound is computed as , where is the largest integer that satisfies the inequality

     

with . Likewise, if you specify CIPCTLDF(TYPE = UPPER), a one-sided lower confidence bound is computed as , where is the largest integer that satisfies the inequality

     

Note that confidence limits for percentiles are not computed when a WEIGHT statement is specified. See Example 4.10.

Previous Page | Next Page | Top of Page