IDAX.T_PMD_TEST - t-Student test for the expected mean difference between two paired columns

The T_PMD_TEST stored procedure calculates the t-Student statistics of two paired numeric input columns. The goal is to evaluate the significance of the difference of their mean values that are compared to an expected value.

Syntax

IDAX.T_PMD_TEST(in parameter_string varchar(32672))

Parameter descriptions

parameter_string
Mandatory one-string parameter that contains pairs of <parameter>=<value> entries that are separated by a comma.
Data type: VARCHAR(32672)
The following list shows the parameter values:
intable
Mandatory.
The name of the input table.
Data type: VARCHAR(256)
outtable
Mandatory.
The name of the output table to which the t-Student statistics are written.
The output table contains the TSTUDPAIREDMEANDIFF column and the PERCENTAGE column.
If the by parameter is specified, an extra BY column is added to indicate for which group the t-Student statistics are calculated.
The TSTUDPAIREDMEANDIFF column contains a string with the t-Student value and the degree of freedom.
The PERCENTAGE column indicates the probability that the column has the expected mean value:
  • If the percentage ranges from 0 to 0.05, the Y column has significantly bigger values than the X column.
  • If the percentage ranges between 0.05 and 0.95, the expected mean difference value matches the difference between the mean values of the Y column and the X column.
  • If the percentage ranges between 0.95 and 1, the Y column has significantly smaller values than the X column.
Data type: VARCHAR(256)
incolumn
Mandatory.
The two numeric columns of the input table that are separated by a semicolon (;).
One column must be followed by :X, the other column must be followed by :Y.
Data type: VARCHAR(ANY)
expdiff
Optional.
The expected difference between the mean values of the input columns (Y - X).
Default: 0
Data type: DOUBLE
by
Optional.
The column of the input table that splits the data into groups for which the stored procedure is to be done.
Default: none
Data type: VARCHAR(128)

Returned information

DOUBLE the probability that the two columns of the input table have different values for the expdiff parameter on average, or the number of groups in the output table for which t-Student statistics are calculated.

Example

CALL IDAX.T_PMD_TEST('intable=IRIS,incolumn=petallength:X;sepallength:Y, outtable=PMD_TEST_RESULT1');
CALL IDAX.T_PMD_TEST('intable=IRIS, incolumn=petallength:X;sepallength:Y, outtable=PMD_TEST_RESULT2, by=class');