1.1 - 8.10 - LAR Syntax Elements - Teradata Vantage

Teradata Vantage™ - Machine Learning Engine Analytic Function Reference

Product
Teradata Vantage
Release Number
1.1
8.10
Published
October 2019
Content Type
Programming Reference
Publication ID
B700-4003-079K
Language
English (United States)
Last Update
2019-12-31
OutputTable
Specify the name of the output table.
TargetColumns
Specify the names of the columns of the InputTable that contain the response and predictors.

This is the syntax of predictor_columns:

{col[,...] | [start_column:end_column]}[,...]

where col is a column name and start_column and end_column are the column indexes of the first and last columns in a range of columns. The range includes start_column and end_column.

The leftmost column has column index 0, the column to its immediate right has column index 1, and so on.

  • In a column range, brackets do not indicate optional elements. You must include the bracket characters (for example, '[2:6]').
  • This function can take at most 799 response and predictor variables.
FitMethod
[Optional] Specify the method to use for linear regression.
Default: 'lasso'
Intercept
[Optional] Specify whether an intercept is included in the model (and not penalized).
Default: 'true'
L2Normalization
[Optional] Specify whether each predictor is standardized to have unit L2 norm.
Default: 'true'
MaxIterNum
[Optional] Specify the maximum number of steps the function runs.
Default: 8*min(number_of_predictors, sample_size - intercept). For example, if the number of predictors is 11, the sample size (number of rows in the InputTable) is 1532, and the intercept is 1, then the default value is 8*min(11, 1532 - 1) = 88.