CORRELATION: Calculating the Degree of Correlation Between Two Sets of Data

The CORRELATION function calculates the correlation coefficient between two numeric fields. The function returns a numeric value between zero (-1.0) and 1.0.

Calculate the Correlation Coefficient Between Two Fields

CORRELATION(field1, field2)

where:

field1

Numeric

Is the first set of data for the correlation.

field2

Numeric

Is the second set of data for the correlation.

Note: Arguments for CORRELATION cannot be prefixed fields. If you need to work with fields that have a prefix operator applied, apply the prefix operators to the fields in COMPUTE commands and save the results in a HOLD file. Then, run the correlation against the HOLD file.

Calculating a Correlation

The following request calculates the correlation between the DOLLARS and BUDDOLLARS fields converted to double precision.

DEFINE FILE ibisamp/ggsales

DOLLARS/D12.2 = DOLLARS;

BUDDOLLARS/D12.2 = BUDDOLLARS;

END

TABLE FILE ibisamp/ggsales

SUM DOLLARS BUDDOLLARS 

CORRELATION(DOLLARS, BUDDOLLARS)

ON TABLE SET PAGE NOLEAD

ON TABLE SET STYLE *

GRID=OFF,$

ENDSTYLE

END

The output is shown in the following image.

CORRELATION calculates the correlation between DOLLARS and BUDDOLLARS.

CORRELATION(DOLLARS, BUDDOLLARS)

For DOLLARS=46,156,290.00 and BUDDOLLARS=46,220,778.00, the result is 0.895691073.