Time Series Model Package

IDMSPEC Methods

IDMSPEC.Close Method

  • rc=obj.Close ();

Finalizes the IDM model in the IDMSPEC object and prepares the IDM model for use in a TSM object or to be imported to a TSMSPEC object for printing or storage to a model repository catalog.

Arguments

There are no arguments associated with this method.

IDMSPEC.Open Method

  • rc=obj.Open ();

Opens the IDMSPEC object for configuration and initializes it with default options. The default options specify that the best IDM model from among the average demand and Croston’s model be selected based on the RMSE criterion (see the value BEST for the 'MODEL' argument in the SetOption method). The smoothing model for the individual components of both models is selected from among the best candidate nonseasonal smoothing method (see value BESTN for the 'Method' argument in the SetMethod method). In addition, by default, all component transformations are disabled (see the value NONE for the 'Type' argument in the SetTransform method).

Arguments

There are no arguments associated with this method.

IDMSPEC.SetMethod Method

  • rc=obj.SetMethod ('IDMComp', 'Method');

Sets a smoothing method for IDM model component.

Input Arguments

You must specify the following input arguments:

'IDMComp'

is a character string that specifies the IDM component model. You can specify one of the following values:

AVERAGE

specifies the average demand model.

INTERVAL

specifies demand interval model.

SIZE

specifies demand size model.

'Method'

takes a string value that specifies the name of the smoothing method. You can specify one of the following values:

BESTN

requests the best candidate nonseasonal smoothing model among the SIMPLE, LINEAR, or DAMPTREND methods.

DAMPTREND

requests damped trend exponential smoothing.

LINEAR

requests linear (Holt) exponential smoothing.

NONE

disables the use of a smoothing model for the specified component.

SIMPLE

requests simple (single) exponential smoothing.

The default is BESTN.

IDMSPEC.SetOption Method

  • rc=obj.SetOption ('Name', Value <,'Name',Value,...>);

Specifies the options for the IDMSPEC object.

Input Arguments

You must specify at least one of the following 'Names' and its associated Value:

'BASE'

takes a numeric Value that specifies the base demand level. A 'BASE' value that is specified as a missing numeric value is interpreted to infer the base demand automatically as the median value of the dependent series.

'CRITERION'

takes a string Value that specifies the model selection criterion (statistic of fit) to be used to select from several candidate models. For a list of valid values, see the CRITERION= option in the HPFDIAGNOSE procedure in SAS Forecast Server Procedures: User's Guide. The default is RMSE.

'MODEL'

is a character string that specifies a mnemonic for the IDM model to use. You can specify the following Values:

AVERAGE

uses the single smoothing model to fit the average demand component.

BEST

requests the best IDM model from among the AVERAGE and CROSTON models.

CROSTON

uses the two smoothing models to fit the demand interval component and the demand size component.

The default is BEST.

IDMSPEC.SetParm Method

  • rc=obj.SetParm ('IDMComp', 'ESMComp',Parm <,Noest,LRest, URest>);

Sets parameter value and restrictions for the ESM component in the IDM component model. The ESM method-specific bounds are employed to limit or filter the smoothing method weight values.

Input Arguments

You must specify the following input arguments:

'IDMComp'

is a character string that specifies the IDM component model. You can specify one of the following values:

AVERAGE

specifies the average demand model.

INTERVAL

specifies demand interval model.

SIZE

specifies demand size model.

'ESMComp'

is a character string that specifies the ESM component of the IDM component. You can specify one of the following values:

DAMP

specifies that Parm is the initial value of a damping weight parameter.

LEVEL

specifies that Parm is the initial value of a level weight parameter.

SEASON

specifies that Parm is the initial value of a season weight parameter.

TREND

specifies that Parm is the initial value of a trend weight parameter.

Parm

takes a numeric value that specifies the smoothing component weight.

You can also specify the following input arguments:

Noest

takes a Boolean value (0 or 1) that, when set to 1, specifies that the Parm argument is fixed. The default value is 0 (initial).

LRest

takes a numeric value between –1 and 2 that specifies a lower bound restriction on the ESM weight.

URest

takes a numeric value between –1 and 2 that specifies an upper bound restriction on the ESM weight.

IDMSPEC.SetTransform Method

  • rc=obj.SetTransform ('IDMComp', 'Type' <,'Option',Parm>);

Sets the functional transform to be used by the IDM model component.

Input Arguments

You must specify the following input arguments:

'IDMComp'

is a character string that specifies the IDM component model. You can specify one of the following values:

AVERAGE

specifies the average demand model.

INTERVAL

specifies demand interval model.

SIZE

specifies demand size model.

'Type'

takes a string value that specifies the transform to use. You can specify one of the following values:

AUTO

automatically chooses between NONE and LOG based on model selection criteria.

BOXCOX(value)

requests Box-Cox transformation with a parameter value between –5 and 5. The default is BOXCOX(1).

LOG

requests logarithmic transformation.

LOGIT

requests logistic transformation.

NONE

does not apply a transformation.

SQRT

requests square-root transformation.

The default is NONE.

You can also specify the following input arguments:

'Option'

takes a string value that specifies prediction semantics for the inverse transform. You can specify the following values:

MEAN

requests that the inverse transform produce mean forecasts.

MEDIAN

requests that the inverse transform produce median forecasts.

The default is MEAN.

Parm

takes a numeric value between –5 and 5 that specifies a control parameter. This parameter is allowed only for Box-Cox transforms.

Last updated: June 14, 2018