SED navigation bar go to SED home page go to Dataplot home page go to NIST home page SED Home Page SED Staff SED Projects SED Products and Publications Search SED Pages
Dataplot Vol 2 Vol 1

WINSORIZED CORRELATION

Name:
    WINSORIZED CORRELATION (LET)
Type:
    Let Subcommand
Purpose:
    Compute the Winsorized correlation of a variable.
Description:
    The standard correlation estimate can be heavily influenced by extreme values. The Winsorized correlation compensates for this by setting the tail values equal to a certain percentile value. For example, for a 90% Winsorized correlation, the bottom 5% of the values are set equal to the value corresponding to the 5th percentile while the upper 5% of the values are set equal to the value corresponding to the 95th percentile. Then the standard correlation formula (HELP CORRELATION for details) is applied.
Syntax:
    LET <par> = WINSORIZED CORRELATION <y1> <y2>        <SUBSET/EXCEPT/FOR qualification>
    where <y1> is the first response variable;
                  <y2> is the second response variable;
                  <par> is a parameter where the computed Winsorized correlation is stored;
    and where the <SUBSET/EXCEPT/FOR qualification> is optional.
Examples:
    LET A = WINSORIZED CORRELATION Y1 Y2
    LET A = WINSORIZED CORRELATION Y1 Y2 SUBSET TAG > 2
Note:
    The analyst must specify the percentages to Winsorize in each tail. This is done by defining the internal variables P1 (the lower tail) and P2 (the upper tail). For example, to Winsorize 10% of each tail, do the following:

      LET P1 = 10
      LET P2 = 10
      LET A = WINSORIZED CORRELATION Y1 Y2
Note:
    Support for the Winsorized correlation has been added to the following plots and commands:

      WINSORIZED CORRELATION PLOT
      BOOTSTRAP WINSORIZED CORRELATION PLOT
      JACKNIFE WINSORIZED CORRELATION PLOT
      CROSS TABULATE WINSORIZED CORRELATION PLOT
      WINSORIZED CORRELATION INTERACTION STATISTIC PLOT

    In addition, the command

      SET CORRELATION TYPE WINSOR

    can be used to specify that the command

      LET C = CORRELATION MATRIX M

    generates Winsorized correlations for a matrix (instead of the standard correlation coefficient normally generated).

Default:
    None
Synonyms:
    None
Related Commands:
    WINSORIZE = Winsorize a variable.
    WINSORIZED STANDARD DEVIATION = Compute a Winsorized standard deviation.
    WINSORIZED MEAN = Compute a Winsorized mean.
    WINSORIZED VARIANCE = Compute a Winsorized covariance.
    WINSORIZED COVARIANCE = Compute a Winsorized covariance.
    VARIANCE = Compute the variance.
    BIWEIGHT MIDCOVARIANCE = Compute the biweight midcovariance.
Applications:
    Robust Data Analysis
Implementation Date:
    7/2002
Program: 
    DIMENSION 20 COLUMNS 
SKIP 25 
COLUMN LIMITS 20 80 
READ AUTO79.DAT Y1 TO Y12 
LET A1 = CORRELATION Y1 Y2 
LET A2 = WINSORIZED CORRELATION Y1 Y2

Date created: 7/22/2002
Last updated: 4/4/2003
Please email comments on this WWW page to alan.heckert@nist.gov.