SAS/ETS 9.22 User''''s Guide 295 pdf

2932 F Chapter 47: SAS/ETS Model Editor Window ReferenceEnd: specifies the ending date for the time series in the data set.. mapping of the variables from the model template to the input

2932 F Chapter 47: SAS/ETS Model Editor Window Reference

End:

specifies the ending date for the time series in the data set Enter a date value in this field, using a form that is recognizable as a SAS date informat (for example, 1998:1, feb1997, or 03mar1998)

SAS Code

provides the SAS statements for the model

Add button

adds a variable row to the table

Delete button

removes variable row from the table

OK

saves the model template

Cancel

closes the window without implementing any changes

Help

provides help about this window

Constraints Details

Use this window to specify the boundary constraints and linear and nonlinear restrictions for the parameter estimates To open this window, right-click a SAS library in the SAS libraries panel, and then select New Model Template from the resulting menu, and then select click Constraints in the left pane of the window

Figure 47.4 Create New Model Template—Constraints

2934 F Chapter 47: SAS/ETS Model Editor Window Reference

Figure 47.4 continued

When Constraints is selected on the Details tab, the following controls and fields are displayed: Simple Bounds

specifies the boundary constraints To create a new row, click the Add button A new entry appears in the Simples Bounds table

To edit the value for an entry, double-click the appropriate row To delete an entry, select an existing entry and then click the Delete button

Restrictions

lists the linear and nonlinear restrictions To create a new row, click the Add button A new entry appears in the Restrictions table

To edit the values for an entry, double-click the appropriate cell in the Simple Bounds table

To delete an entry, select an existing entry and then click the Delete button

New Fitted Model Wizard

The New Fitted Model wizard provides a step-by-step interface that enables you to create and define the equation statements, variable definitions, parameter definitions, and constraints that can be used to

fit many different market data sets You can create a fitted model either from scratch or by modifying

an existing model template The following instructions walk you through creating a fitted model from an existing model template To create a new fitted model from an existing model template, right-click a named model template in the SAS libraries panel, and then select New Fitted Model The Fitted Model wizard opens

The Fitted Model wizard sequentially asks you to provide the following information:

1 name of your model and the location where it is to be stored

2 name of an existing input data set

3 creation or verification of the formula for your model

4 model parameters and variables

5 mapping of the variables from the model template to the input data set variables if the fitted model is created using a model template

6 name of any output data sets desired (optional)

7 configuration of the fit options (such as estimation methods and optimization settings) for the model

After you have provided the preceding information, the new model is fitted against the supplied input data set, and the results and model are stored at the location specified If the new fitted model is built from a model template, its content is similar to the template content with the following exceptions: the input and output data sets are defined and all of the model program variables are changed based

on the variable mapping After a fitted model is created, subsequent changes that are made to the model template from which it was created have no effect on the fitted model

The lower part of each page in the Fitted Model wizard contains the following controls:

 Details provides the options for each step

 SAS Code provides the SAS statements for the entire model

 Back moves the wizard back to the previous page

 Next moves the wizard forward to the next page

 Finish saves the model and causes the model program to run and generate the statistical results

 Cancel closes the wizard unconditionally, and nothing is saved

 Help provides help about the wizard

The following sections show the step-by-step process of the Fitted New Model wizards

2936 F Chapter 47: SAS/ETS Model Editor Window Reference

Name Your Model Page

The initial page of the wizard is the Name Your Model page This page informs you that you are creating a fitted model from a model template You can change the name of the model You must specify the library and data set where your model is to reside The name must be unique within the library and data set CAUTION: If you do not change the name or library, the original model template is overwritten

Figure 47.5 Model Fitting Wizard

This page has the following controls and fields:

Name:

is the name the current new model

Description:

is the description of the model

Library:

is the SAS library where the model is to be stored

Data Set:

is the SAS data set where the model and fitted results are to be stored

Select the Data Set to Fit Page

The Select the Data Set to fit page asks you to specify the name and location of an input data set You can also view time series of any selected variables

Figure 47.6 Model Fitting Wizard

This page has the following controls and fields:

Library Name:

is the library where the input data set is located

Data Set Name:

is the input data set against which you want the model fitted

Browse

opens the data set selection window for selecting an input data set

After you specify the input data set, the variable names in the data set are listed in the Selected Variables to View Time Series table

Check Data Time Series

is the list of variables, ID, frequency, and time series property of the input data set

Select Variables to View Time Series

lists the name and label for each variable in the data set

ID:

specifies a single variable to identify observations in error messages, in other listings, or in the output data set The ID variables are usually SAS date or datetime variables You can type any valid SAS variable name in this field or select from the list

Frequency:

is the time interval (data frequency) for the input data set

Specify

opens the Specify Frequency window See the section “Specify Frequency Window” on page 2938 for more information

View Time Series

opens a window that displays a time series plot and of the table of observations for the selected

2938 F Chapter 47: SAS/ETS Model Editor Window Reference

variables series See the section “View Time Series of the Data Window” on page 2939 for more information

Specify Frequency Window

When you click Specify in the Select the data set to fit page, the Specify Frequency dialog box opens This dialog box enables you to specify the name of the time variables This dialog box is equivalent

to the INTERVAL= option in the ID statement in PROC MODEL See Chapter 18, “The MODEL Procedure” (SAS/ETS User’s Guide), for more information about PROC MODEL statements and options

Figure 47.7 Specify Frequency Window

This dialog box has the following controls and fields:

Type:

specifies the type of the date values For example, Day specifies date values that correspond to single day periods Hour specifies datetime values that correspond to single hour periods Multiplier:

specifies a multiplier for date values The value 1 selects date values that correspond to periods of one of the periods specified by the Type value The value 2 selects date values that correspond to periods that are twice the duration specified by the Type value

Shift:

specifies the date values shift The value 1 selects date values that correspond to periods of one

date value beginning on the first date value The value 2 selects date values that correspond to periods of one date value beginning on the second date value

Weekend:

specifies days to be considered as weekends

Date values include time of day

controls whether the date values include time of day When this check box is selected, date values include the time of day

OK

closes the window and returns to the Select the dataset to fit page

Cancel

closes the window without implementing any changes and returns to the Select the dataset to fit page

View Time Series of the Data Window

Figure 47.8 View Time Series of the Data

2940 F Chapter 47: SAS/ETS Model Editor Window Reference

Figure 47.8 continued

This window has the following controls and fields:

Graphing Options

opens a window that enables you to change graphing options

Series Values

displays a plot of the time series of the selected variables

Observation

displays a table of all the observations of the data set

OK

closes the View Time Series window

Assign Variables Page

This page enables you to assign variables from the input data set to different categories such as endogenous variables, exogenous variables, and instruments variables

Figure 47.9 Model Fitting Wizard

This page has the following controls and fields:

Select Variables to Assign table

lists the variable names and labels of the input data set

Endogenous Variables: table

lists the endogenous (dependent) variables of the model An endogenous variable is a variable that is determined within the model Endogenous variables can appear on the right and left sides of model equations, but usually only on one side for each equation

Exogenous Variables: table

lists the exogenous (independent) variables of this model An exogenous variable is a variable that is determined outside of the model Exogenous variables appear on the right side of model equations, and by default exogenous variables are used as instruments for estimation methods that require instrumental variables (for example, 2SLS, IT2SLS, 3SLS, IT3SLS, GMM, and ITGMM)

Instruments: table

lists the model instrumental variables Instrumental variables are used in estimation methods that require instrumental variables (for example, 2SLS, IT2SLS, 3SLS, IT3SLS, GMM, and ITGMM)

Instrumental variable estimation methods are appropriate when a model contains a right-side regressor variable that is correlated with the error term In 2SLS and 3SLS, a first-stage regression is performed to create a linear combination of instruments to replace the regressor that is correlated with the error term Appropriate instrumental variables are correlated with the regressor they are replacing, but they are uncorrelated with the error term They should not depend on the endogenous variables

Right Arrow button

enables you to move highlighted variables into the table to the right of the button

Delete button

enables you to delete highlighted variables from the table to the left of the button