7.00.02 - Output - Aster Analytics

Teradata Aster® Analytics Foundation User GuideUpdate 2

Product
Aster Analytics
Release Number
7.00.02
Published
September 2017
Content Type
Programming Reference
User Guide
Publication ID
B700-1022-700K
Language
English (United States)
Last Update
2018-04-17
The Arima function has two output tables:
  • Model table

    The model table contains the coefficients of the model. The function outputs the coefficients to both the model table and the console.

  • [Optional] Residual table

    The residual table contains the value and residual for each time point.

ARIMA Model Table Schema
Column Name Data Type Description
coef VARCHAR Contains the coefficient names in the left column of the following table.
value VARCHAR Contains the coefficient values in the right column of the following table.
Arima Model Coefficients
Coefficient Name Value
coef Vector of the coefficients p, d, q, sp, sd, and sq.
ar_params Vector of the auto regression parameters that correspond to the coefficients
ar_params_sd Vector of the standard deviations that correspond to the auto regression parameters
ma_params Vector of the moving average parameters that correspond to the coefficients
ma_params_sd Vector of the standard deviations that correspond to the moving average parameters
seasonal_ar_params Vector of the seasonal auto regressive parameters that correspond to the coefficients
seasonal_ar_params_sd Vector of the seasonal standard deviations that correspond to the auto regressive parameters
seasonal_ma_params Vector of the seasonal moving average parameters that correspond to the coefficients
seasonal_ma_params_sd Vector of the seasonal standard deviations that correspond to the moving average parameters
mean_param Mean value (This row appears if you specify IncludeMean('true').)
mean_param_sd Standard deviation for the mean value (This row appears if you specify IncludeMean('true').)
period Period of a season (For the nonseasonal model, the period is 0.)
sigma2 Variance
loglikelihood Partial log-likelihood
iterations Iteration that the function ran
converged 'true' if the training is converged, otherwise 'false'
ARIMA Residual Table Schema
Column Name Data Type Description
timestamp_column Any The table can have more than one such column. These columns contain the sequence (time points) of the input parameters.
value Same type as value_column in input_table Contains the values of the time points.
residual DOUBLE PRECISION Contains the residuals of the time points.