7.00.02 - Output - Aster Analytics

Teradata Aster® Analytics Foundation User GuideUpdate 2

Product
Aster Analytics
Release Number
7.00.02
Release Date
September 2017
Content Type
Programming Reference
User Guide
Publication ID
B700-1022-700K
Language
English (United States)

The output of VARMAX is a model for each partition in the input table. The Output table schema and additional details about the output are shown in the following two tables.

VARMAX Output Schema
Column Name Data Type Description
partition_columns Any Identifies the partition_column from the input table.
coef VARCHAR See the following table for the values that appear in this column.
coef_value VARCHAR See the following table for explanations of the values that appear in this column.
stepahead Integer Included only if the StepAhead argument has a positive value.

Identifier of the future period. For example, if the argument StepAhead is 3, this column shows 1, 2, 3 to indicate the future period for which a predicted value is shown in the next column.

predict_columnname Double Included only if the StepAhead argument has a positive value. One column appears for each column specified in the ResponseColumns argument.

The predicted future values of the response variable.

VARMAX Model Coefficients
coef coef_value
coef The vector [p, d, q, sp, sd, sq, b].
ar_params The matrices Φi, shown as a vector of p matrices, each of which is an n * n matrix. p is from the coef vector and n is the number of response variables specified in the ResponseColumns argument.
ma_params The matrices Φi, shown as a vector of q matrices, each of which is an n * n matrix. q is from the coef vector and n is the number of response variables specified in the ResponseColumns argument.
exogenous_params The matrices B i , shown as a vector of b matrices, each of which is an n * m matrix. b is from the coef vector, n is the number of response variables specified in the ResponseColumns argument, and m is the number of exogenous variables specified in the ExogenousColumns argument.
seasonal_ar_params The matrix Φsi for the seasonal parameters.
seasonal_ma_params The matrix Φsi for the seasonal parameters.
mean_param The mean vector of the response series. This value is only displayed if the argument IncludeMean is set to True.
period The cycle period for seasonal models (0 for non-seasonal models).
lag The lag value specified in the function call.
sigma The variance matrix.
aic The Akaike information criterion.
bic The Bayesian information criterion.
iterations The number of iterations performed.
converged Whether the algorithm converged.