TD_HOLT_WINTERS_FORECASTER Syntax Elements - 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
lifecycle
latest
SERIES_SPEC
Time series passed into the TD_HOLT_WINTERS_FORECASTER function as an input. Input values must be of content type REAL.

See Series Specifications.

FUNC_PARAMS
Name Data Type Description
FORECAST_PERIODS Integer Number of periods to forecast. Must be greater than 0.
ALPHA FLOAT [Optional] The ALPHA parameter controls the smoothing relative to the level component of the forecasting equation. If present, this value is used in the equation to perform the forecasting. Value must be greater than 0 and less than 1.

If not present, then the ALPHA value is estimated using goodness-of-fit metrics.

BETA FLOAT [Optional] The BETA parameter controls the smoothing relative to the trend component of the forecasting equation. If present, this value is used in the equation to perform the forecasting. Value must be greater than 0 and less than 1.

If not present, then the BETA value is estimated using goodness-of-fit metrics.

GAMMA FLOAT

[Dependencies: If the value is greater than zero, then this parameter is only used when the SEASONAL_PERIODS parameter has a value greater than one. The value must be greater than or equal to 0 and less than or equal to 1.]

The GAMMA parameter controls the smoothing relative to the seasonal component of the forecasting equation. If present, this value is used in the equation to perform the forecasting.

If not present and the SEASONAL_PERIODS parameter is used, then the GAMMA value is estimated using goodness-of-fit metrics.

SEASONAL_PERIODS Integer [Dependencies: The parameter value must be greater than 1 if the GAMMA parameter is used.] Number of periods or sample points for one season. For example, yearly data with monthly sample points, the parameter is 12, and quarterly with monthly sample points, the parameter is 4.

By default, the parameter is not used in calculations.

INIT(LEVEL_0()) FLOAT [Optional] If the LEVEL_0 value is provided, the initialization value is used as part of the fitting and forecasting operations.

If the SEASONAL_PERIODS parameter is used, then the default value is an average of the first set of input series values up to the number of seasonal periods. If the parameter is not used, then the default is the second value of the input series.

INIT(TREND_0()) FLOAT [Optional] If the TREND_0 value is provided, the initialization value is used as part of the fitting and forecasting operations. If a TREND_0 value is not provided, then the initialization value is calculated as an additive trend.

The default value is calculated based on the input series values and the parameters values of SEASONAL_PERIODS and MODEL.

INIT(SEASON_0()) FLOAT [Dependencies: The SEASONAL_PERIODS parameter must be specified when this parameter is used.] A list of values, one per period. If the SEASON_0 value is provided, the initialization value is used as part of the fitting and forecasting operations. If a SEASON_0 value is not provided, then the initialization value is calculated as a multiplicative seasonality.

The default value is calculated based on the parameters values for SEASONAL_PERIODS and MODEL, and the input series values.

MODEL Enumerated type Type of Holt Winters forecasting. Options are ADDITIVE or MULTIPLICATIVE.

Controls the level and seasonality. The trend remains additive in either case. The default is ADDITIVE.

FIT_PERCENTAGE Integer [Optional] The value must be greater than 0 and less than or equal to 100. Percentage of passed-in sample points to use for the model fitting or parameter estimation. The default is 100.
PREDICTION_INTERVALS Enum, String [Optional] The confidence level for the prediction, such that 80 means 80% confidence.
  • NONE
  • 80
  • 90
  • BOTH 80 and 95

Default is BOTH confidence intervals are generated.

FIT_METRICS Integer [Optional] Flag to generate the secondary result set that contains the model metadata statistics. A value of 1 means generate the secondary result set. A value of 0 means do not generate the secondary result set. The default is 0.

The generated result set is retrieved using TD_EXTRACT_RESULTS against the analytical result table containing the results.

SELECTION_METRICS Integer [Optional] Flag to generate the tertiary result set that contains the selection metrics. A value of 1 means generate the tertiary result set. A value of 0 means do not generate the tertiary result set. Default is 0.

The generated result set can be retrieved by issuing a TD_EXTRACT_RESULTS against the analytical result table containing the results.

RESIDUALS Integer [Optional] Flag to generate the quaternary result set that contains the model residuals. A value of 1 means generate the quaternary result set. A value of 0 means do not generate the quaternary result set. The default value is 0.

The generated result set is retrieved using TD_EXTRACT_RESULTS against the analytical result table containing the results.

INPUT_FMT
No INPUT_FMT options are available for this function.
OUTPUT_FMT
[Optional] Specify the INDEX_STYLE of the output format. Options are NUMERICAL_SEQUENCE or FLOW_THROUGH. The default is NUMERICAL_SEQUENCE.