TD_HOLT_WINTERS_FORECASTER Output - Teradata Vantage

Database Unbounded Array Framework Time Series Functions
Deployment
VantageCloud
VantageCore
Edition
Enterprise
IntelliFlex
VMware
Product
Teradata Vantage
Release Number
17.20
Published
June 2022
Language
English (United States)
Last Update
2024-10-04
dita:mapPath
ncd1634149624743.ditamap
dita:ditavalPath
ruu1634160136230.ditaval
dita:id
ncd1634149624743
The TD_HOLT_WINTERS_FORECASTER function produces a multilayer ART and returns up to four result sets.
  • Primary (ARTPRIMARY) result set contains the historical data points, the forecasted data points, and the prediction level data points. This result set is retrieved using a SELECT on the analytical result table.
  • Secondary (ARTFITMETADATA) result set contains a collection of goodness-of-fit metrics similar to those returned by the TD_FIT_METRICS function.
  • Tertiary (ARTSELMETRICS) result set contains a collection of model selection or validation metrics similar to those returned by the TD_SELECTION_CRITERIA function.
  • Quaternary (ARTFITRESIDUALS) result set contains the residual series associated with the fitting and forecasting operation.

RETURNS TABLE Schema for the Primary Output

Name Data Type Description
derived-series-identifier Varies The resultant series identifier inherited from the SERIES_ID.
ROW_I Varies

Data depends on OUTPUT_FMT specification.

 Index of the series.
OBSERVED_VALUE FLOAT Original observed (historical) sample values associated with the series.
FORECAST_VALUE FLOAT Forecasted values for the model.
LO_80 FLOAT Low end of the 80% prediction interval.
HI_80 FLOAT High end of the 80% prediction interval.
LO_95 FLOAT Low end of the 95% prediction interval.
HI_95 FLOAT High end of the 95% prediction interval.

RETURNS TABLE for the Secondary Result Set

This result set is located in LAYER1 under the name ARTFITMETADATA. These results can be retrieved by issuing a TD_EXTRACT_RESULTS against the ART.

Name Data Type Description
derived-series-identifier Varies The resultant series identifier inherited from the SERIES_ID.
NUM_SAMPLES Integer The number of sample points used to fit the model.
VAR_COUNT Integer The number of explanatory variables, including the constant, in the original regression.
R_SQUARE FLOAT The calculated R-squared value from the original and calculated values.
R_ADJ_SQUARE FLOAT The calculated adjusted R-squared value from the original and calculated values.
STD_ERROR FLOAT The standard error or deviation associated with the model.
STD_ERROR_DF FLOAT The degrees of freedom associated with the standard error calculation.
ALPHA FLOAT The base value.
BETA FLOAT The trend value.
GAMMA FLOAT The seasonal component.
ME FLOAT The Mean Error.
MSE FLOAT The calculated Mean Square Error.
MAE FLOAT The Mean Absolute Error.
MPE FLOAT The Mean Percent Error.
MAPE FLOAT The Mean Absolute Percent Error.
FSTAT_CALC FLOAT The calculated F-statistic value for the OLS regression.
P_VALUE FLOAT The p-value corresponding to the calculated test statistic.
NUM_DF Integer The degrees of freedom in the numerator associated with the unexplained portion of the F-statistic.
DENOM_DF Integer The degrees of freedom in the denominator associated with the explained portion of the F-statistic.
SIGNIFICANCE_LEVEL FLOAT Level of significance for the test.
F_CRITICAL FLOAT The chi-squared critical value extracted from the chi-squared-statistic tables.
F_CRITICAL_P FLOAT The p-value corresponding to the calculated critical value.
NULL_HYPOTH VARCHAR(16) The result of the F-test.
  • ACCEPT means that the null hypothesis is accepted, and the slope coefficients are zero.
  • REJECT means that the null hypothesis is rejected, and there are slope coefficients that are not zero.

RETURNS TABLE for the Tertiary Result Set

These results can be retrieved by issuing a TD_EXTRACT_RESULTS against the ART.

Name Data Type Description
derived-series-identifier Varies The resultant series identifier inherited from the SERIES_ID.
NUM_SAMPLES Integer The number of sample points used to fit the model.
VAR_COUNT Integer Total number of parameters involved in the model.

For an ARMA(p,q) model, the calculation of VAR_COUNT is p + q + 1.
AIC FLOAT The calculated Akaike Information Criteria value.
SBIC FLOAT The calculated Schwarz Bayesian Information Criteria value.
HQIC FLOAT The calculated Hannon Quinn Information Criteria value.
MLR FLOAT The calculated Maximum Likelihood Rule value.
MSE FLOAT The calculated Mean Square Error value.

RETURNS Table Schema for Quaternary Result Set

These results can be retrieved by issuing a TD_EXTRACT_RESULTS against the ART.

Name Data Type Description
derived-series-identifier Varies The resultant series identifier inherited from the SERIES_ID.
ROW_I Integer Indexing column for the one-dimensional multivariate output array containing the residuals. This could be integer or flow through type given by input series.
ACTUAL_VALUE FLOAT The actual value of the response variable.
CALC_LEVEL FLOAT The calculated value of the level component.
CALC_TREND FLOAT The calculated value of the trend component.
CALC_SEASONAL FLOAT The calculated value of the seasonal component.
CALC_VALUE FLOAT The calculated value of the response variable using the model.
RESIDUAL FLOAT The difference between the calculated response value and the actual response value.