Optional Parameters - Teradata Warehouse Miner

In-Database Analytic Functions User Guide

Product
Teradata Warehouse Miner
Release Number
5.4.4
Published
August 2017
Language
English (United States)
Last Update
2018-05-04
dita:mapPath
guj1484331868727.ditamap
dita:ditavalPath
ft:empty
dita:id
B035-2306
lifecycle
previous
Product Category
Teradata® Warehouse Miner
outputdatabase
The database that will contain the output score table.
If outputdatabase and outputtablename are not both specified, a volatile output table with randomly generated name is created in the logon user database.
outputtablename
The name of the score output table containing key columns and predicted values of the dependent variable in the linear model. The output table may also contain retained columns passed through from the input to the output table unchanged, as well as a residual value containing the difference between the actual and predicted values of the dependent variable column. The output table may also contain group by columns if these are present in the model table.

Note that if the output table already exists it must first be dropped by the user if outputdatabase and outputtablename are both specified. If outputdatabase and outputtablename are not both specified because only model evaluation is being performed, a volatile output table with a randomly generated name is created in the logon user database, and the output result set is returned to the user instead.

index
By default, the primary index columns of the score output table are the primary index columns of the input table. This parameter allows the user to specify one or more different columns for the primary index of the score output table. Regardless of whether the user uses the default setting or specifies different columns, the index columns are included both in the Primary Index clause and the select list. In addition, the index columns should form a unique key for the score output table. Otherwise, there could be more than one score for a given observation.
retain
One or more columns from the input table can optionally be specified here to be passed along to the score output table.
scoringmethod
Three scoring methods are available as outlined below. By default, the model is scored but not evaluated.
  • Score
  • Evaluate
  • Score and Evaluate
residual
If the score method is score and evaluate, the name of a column that contains the residual value (the difference between the predicted and actual value of the dependent variable) can be given here. By default, this column is named “Residual”.
predicted
If the score method is score or score and evaluate, the name of the predicted value column can be entered here. If not entered here, the name of the dependent column in the input table is used.