The input table consists of an arbitrary number of grouping columns and a single value column that contains the data set to be matched to the specified distributions. The syntax shown includes clauses that create two tables from the input table. One table ranks the data and the other table counts the number of points in each group.
For continuous distributions, if your input table already includes a rank column, replace the clause ON (SELECT RANK()... with the clause ON SELECT * FROM input_table .
| Column Name | Data Type | Description |
|---|---|---|
| column | Any | Column used to partition or identify the values. The table can have several such columns. |
| value_column | INTEGER, BIGINT, NUMERIC, or DOUBLE PRECISION | Contains the values of the sample data set. |