1.1 - 8.10 - PCA Output - Teradata Vantage

Teradata Vantage™ - Machine Learning Engine Analytic Function Reference

Teradata Vantage
Release Number
October 2019
Content Type
Programming Reference
Publication ID
English (United States)

PCAReduce Output Table Schema

Each row represents a principal component. The rows are in descending order based on the standard deviation, which is a measure of the variation in the data set that was captured by that principal component.

Column Data Type Description
component_rank INTEGER Rank of principal component. Components are ranked in descending order of standard deviation (and variance).
dimension_i DOUBLE PRECISION [Column appears once for each dimension.] Values of ith dimension of data set.
sd DOUBLE PRECISION Standard deviation of components in eigenvector represented by row.
var_proportion DOUBLE PRECISION Proportion of variance of components in eigenvector represented by row.
cumulative_var DOUBLE PRECISION Cumulative variance of components in eigenvector represented by row.
mean VARCHAR One row of this column contains a list of average values, one for each target_column_i in the input table. The list has this format:

[average [, average ...]]

The outer brackets appear in the table.

The other rows of this column contain NULL.