SAS/ETS 9.22 User''''s Guide 294 docx

2975 Overview of SAS/ETS Model Editor The SAS/ETS Model Editor enables you to interactively create and edit model templates and fitted models that are associated with econometric time se

2922

Part V SAS/ETS Model Editor (Experimental)

2924

Chapter 47

SAS/ETS Model Editor Window Reference

Contents

Overview of SAS/ETS Model Editor 2926

Invoking the SAS/ETS Model Editor Application 2926

Model Browser Window 2927

Create a New Model Template 2928

Equations Details 2930

Constraints Details 2932

New Fitted Model Wizard 2934

Name Your Model Page 2936

Select the Data Set to Fit Page 2936

Assign Variables Page 2940

Enter (Verify) the Formula for Your Model Page 2942

Map Program Symbols to Data Set Variable Page 2943

Set Fit Options Page 2945

Output and Reports 2952

Equation Results Window 2952

Time Series Window 2954

Graphing Options Window 2955

Model Summary Results Window 2956

Fitted Model Equation Results Window 2958

Fitted Model Covariance/Correlation Matrix Window 2959

Fitted Model Distribution Window 2959

Model Parameters Estimates Window 2960

Residuals Plot Window 2961

Model Summary of Residual Errors Window 2962

Actual v/s Predicted Plot Window 2962

Edit Existing Fitted Model 2963

Open Existing Model 2963

Fit Model—Equations 2964

Fit Model—Input 2965

Fit Model—Method 2967

Fit Model—Iteration 2967

Fit Model—Tests 2969

Fit Model—Results 2970

Define the Model Parameters and Variables 2971

2926 F Chapter 47: SAS/ETS Model Editor Window Reference

Model Equations 2973

Define the Model Parameters, Variables, and Equations 2974

Model Constraints 2975

Overview of SAS/ETS Model Editor

The SAS/ETS Model Editor enables you to interactively create and edit model templates and fitted models that are associated with econometric time series models and risk models These models are developed from time series and cross-sectional data You can use these models to statistically model market behavior

The SAS/ETS Model Editor consists of the following:

 a fitted model wizard with which you can create and define the equation statements, variables and parameters, constraints, and fit options on a step-by-step basis You can apply a fitted model to any specific market data

 a program editor panel, which enables you to write programming code to define your model and create additional dialog boxes The additional dialog boxes enable you to more explicity specify properties associated with your model

 a model template which you can also create to define the equation statements, variables and parameters, and constraints that you need in the programming code Model templates are commonly used models that can be applied to a wide variety of data

You can use either the Fitted Model Wizard or a model template to fit the model to a specific market data set to generate the model parameter estimates and other statistical results as specified in the MODEL procedure The MODEL procedure analyzes models in which the relationships among the variables comprise a system of one or more nonlinear equations Primary uses of the MODEL procedure are estimation, simulation, and forecasting of nonlinear simultaneous equation models See Chapter 18, “The MODEL Procedure” (SAS/ETS User’s Guide), for more information.This chapter provides a reference to the various windows of the SAS/ETS Model Editor The windows are presented in work flow order Each section describes the purpose of the window, how to open it, its controls, fields, and menus

Invoking the SAS/ETS Model Editor Application

You can start the SAS/ETS Model Editor application when using the SAS system interactively by submitting the command%modedit;in the SAS Display Manager Program Editor, as shown in

Figure 47.1

Model Browser Window F 2927

Figure 47.1 Invoking the SAS/ETS Model Editor

Model Browser Window

From the SAS/ETS Model Editor main menu, select View IShow Model Browser to open the Model Browser window The Model Browser window displays a list of the model templates and fitted models along with attributes about each item in the list When the window first opens, the list contains all the model templates and fitted models that are displayed in the model tree in the desktop

To constrain the items in the list, click Find and specify your constraints You can sort the data in each column by clicking the column heading You can also right-click any row in the table to access the same pop-up menu that is available for entries in the model tree on the desktop

Figure 47.2 Model Browser Window

The Model Browser window has the following controls and fields:

2928 F Chapter 47: SAS/ETS Model Editor Window Reference

Types:

specifies which types of model you want to view You can select Model Templates, Fitted Models, or both

Library Name:

is the name of the library (for example, RISKSAMP) where the models are located

NOTE: Predefined libraries, such as RISKSAMP, contain predefined, read-only models You can open and apply these models to specific market data sets But if you want to modify a predefined model template, you need to copy or rename the model template into a temporary directory such as WORK or SASUSER, rename it, and work from there To copy an object to

a desired library, right-click the object and select Duplicate

Data Set Name:

is the input data set in which the model resides

Find

enables you to constrain the data displayed in Model Browser window to only those models that match the search criteria indicated

Model Name:

is the name of the model A model name can contain up to 32 characters, which can be underscores, letters (A–Z or a–z) or numerals (0–9) A model name cannot contain spaces Interval:

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

Specify

specifies which time frequency is to be used in the time series model Click Specify to open the Specify Frequency window, which is described in the section “Specify Frequency Window”

on page 2938

Clear

clears the search criteria that you have currently specified and resets the search fields to the defaults

Create a New Model Template

A model template is a convenient way to predefine a commonly used model After you create a model template, you can use it repeatedly to create a fitted model for various data sets without recreating the model parameters, equations, and constraints To create a new model template, right-click a SAS library in the SAS libraries panel, and then select New Model Template A window opens with its title of the form SASLibrary.newElement

In this window, you can do the following:

 specify the equations and variable and parameter definitions in Equations on the Details tab See “Equations Details” on page 2930 for more details

Create a New Model Template F 2929

 specify the model constraints for the model in Constraints on the Details tab See “ Con-straints Details” on page 2932 for more details

 view the generated SAS code on the SAS Code tab

 check the syntax of the SAS code by clicking the Check Syntax button

 save the model template by clicking OK

For information about equation statements, variable and parameter definitions, and constraints in the MODEL procedure, see Chapter 18, “The MODEL Procedure” (SAS/ETS User’s Guide), for more information

The top of the model window has the following fields:

Name:

is the name of the current new model You can change the default name to another name such

as “MYSPEC” A name can contain up to 32 characters, which can be underscores, letters (A–Z or a–z) or numerals (0–9) A name cannot contain spaces

Description:

is description of the model You can type some description such as “My favorite template” This field is optional

Library:

is the SAS library where the model is to be stored

The model window has two tabs: Details and SAS Code On the Details tab, you can select Equations or Constraints in the left pane; the fields displayed in the right pane change dependent

on the selection in the left pane For more information about the Details tab, see “Equations Details”

on page 2930 and “Constraints Details” on page 2932

When you define the model variables and equations, you can click SAS Code tab to view the MODEL procedure code that is generated by the values specified on the Details tab The following SAS statements provide an example of equivalent PROC MODEL code See Chapter 18, “The MODEL Procedure” (SAS/ETS User’s Guide), for more information about PROC MODEL

proc model outspec=(WORK.MODSPEC specname=CIR speclabel="Cox Ingersoll Ross") ; endogenous rate ;

parms kappa ;

parms theta ;

parms sigma ;

label kappa = 'Speed of Mean Reversion' ;

label theta = 'Long term Mean' ;

label sigma = 'Constant part of Variance' ;

rate = lag(rate) + kappa * (theta - lag(rate) ) ;

h.rate = sigma * sigma * lag(rate);

run;

quit;

2930 F Chapter 47: SAS/ETS Model Editor Window Reference

Equations Details

When you select Equations on the Details tab, you can specify the equation for a model and define the variables and parameters for the model

Figure 47.3 Create New Model Template—Equations

Equations Details F 2931

When Equations is selected on the Details tab, the following controls and fields are displayed:

Equation:

specifies the equations for the model

Variable and Parameter Definitions

lists the variables and parameters for the model that is defined in the Equation: field To add

a new row, click the Add button A new entry appears in the Variable and Parameter Definitions table To edit the value for an entry, double-click the appropriate cell in the Variable and Parameter Definitions table To delete an entry, select an existing entry and then click the Delete button

Name

specifies the name of the variable or parameter

Type

specifies the variable type (such as endogenous)

Label

specifies the label of the variable or parameter

Instrument?

specifies whether the variable or parameter is an instrument

Output To Data Set

specifies the output to data set options The selections map to the options DROP, KEEP, and OUTVARS of the MODEL procedure

All dependent variables are instruments

indicates whether all of the exogenous variables are to be instrumental variables (This check box is equivalent to the _EXOG_ option in the INSTRUMENTS statement in the MODEL procedure See Chapter 18, “The MODEL Procedure” (SAS/ETS User’s Guide), for more information about PROC MODEL

Instruments Only Use Intercept

indicates whether to include only the intercept in the list to be an instrument variable

Instruments Include Intercept

indicates whether to include an intercept term and all of the exogenous variables as instrumental variables

ID

specifies the name of the ID variable to be created for the input data set You can type any valid SAS variable name in this field or select from the list

Range:

specifies the time range between observations in the data set Select a range from list

Start:

specifies the starting 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)