ART_SPEC Element | Teradata Vantage - ART Specifications - Teradata Vantage

The primary use of an analytical result table (ART) is to associate function results with a name label, enabling you to easily retrieve the result data and pass the result to another UAF function. An ART can have multiple layers. Each layer has its own dedicated row composition for the series or matrix. This lets you store and analyze complex data structures in a tabular format. For example, a time series with multiple dimensions can be stored in an ART with multiple layers.

An ART is a table generated by the INTO ART or INTO VOLATILE ART clause of a UAF function. The INTO ART clause creates a permanent table. The INTO VOLATILE ART clause creates a volatile table. A volatile table is only available during the current session and is automatically dropped when the session ends.

You can assign a name to the table and to some of its columns, but all other aspects of the table are set by Teradata. Using EXECUTE FUNCTION INTO VOLATILE ART is slightly faster than using EXECUTE FUNCTION INTO ART. User manipulation of ART tables, such as update, insert and delete, is not allowed. See EXECUTE FUNCTION for information about using INTO ART and INTO VOLATILE ART in Teradata Vantage™ - SQL Data Manipulation Language.

An ART can be referenced by a SERIES_SPEC or ART_SPEC. When referenced by an ART_SPEC, the specification represents a simplified way of referencing the output similar of programming environments such as R and PYTHON. For example, the partial auto-correlation function can take a time series or the output of another auto-correlation function as input.

Similar to R and PYTHON environments, not all time series functions accept a simplified ART_SPEC as input. Time series functions that do accept ART_SPECs have a reference to ART_SPEC as acceptable input in the Syntax section. The following table lists the functions that accept ART_SPECs.

• ART_SPEC Syntax
• ART_SPEC Syntax Elements
• ART_SPEC Examples