Planning a Portfolio

Before you begin a portfolio, you must plan your portfolio resources. Here are questions to consider and conditions to meet for modeling projects within a portfolio:
  • After you know which users are assigned to the projects within a portfolio, a SAS Decision Manager administrator must ensure that the user is assigned to the appropriate user group and role. For more information, see Configuring Users, Groups, and Roles in SAS Decision Manager: Administrator’s Guide.
  • How do you want to structure the projects within the portfolio? A portfolio is an object within a folder. The Portfolios category view enables multiple levels of folders so that you can customize how you structure the portfolios. For more information, see Managing Folders for Model Projects and Portfolios.
  • What models do you want to use in each project of the portfolio? If the models were created using SAS Enterprise Miner, SAS/STAT, or the SAS/ETS procedures COUNTREG and SEVERITY, all model components are available to SAS Decision Manager when you import the model. Only models that are contained in an SPK file can be imported. At least one SPK file must be prepared for each project and the SPK files should be placed in the same location. If your model is a SAS code model or a PMML model that is not contained in an SPK file, you must import it separately into the desired project within the portfolio, after the portfolio is created. You must also ensure that you have imported all of the model component files. For more information, see Import Models from Local Files and Import a PMML Model.
  • What model function do you want to use in each project of the portfolio?
    SAS Decision Manager has several model function types:
    • Classification
    • Prediction
    • Segmentation
    • Analytical
    After the model function is specified for the portfolio, the Model function property for a project cannot be changed. Ensure that the types of models that you are going to use in each project of the portfolio fit within the selected model function type. For more information, see Types of Model Functions.
  • How do you want to define your project input and output variables? When you create a portfolio, you can import the variables using input and output prototype tables. The project variables are set for each project within the portfolio. The prototype tables must be registered in the SAS Metadata Repository. Tables that were registered using the SAS Management Console must also be made available in the Data category view before you create the portfolio. For more information, see Defining Project Input and Output Variables.
  • What method do you want to use to track the progress of a version? The Workflows and My Tasks category views enable you to track the progress of tasks from the version level for each individual project within a portfolio. An authorized user can create a workflow and associate it with a version. For more information, see Overview of Using Workflows.
  • When you publish project champion models from a portfolio to the SAS Metadata Repository, you must specify a location in which to store the models. You might need to create a folder in the SAS Metadata Repository, if one does not already exist. For more information, see Publishing Models from a Portfolio.
  • After your project champion models are in a production environment, you can monitor the performance of the project champion models within a portfolio in SAS Decision Manager using your organization's operational data. If you use SAS Decision Manager to monitor performance of projects within a portfolio, you must first prepare performance tables using the operational data and then register the tables in the SAS Metadata Repository using the Data category view. Tables that are registered to the SAS Metadata Repository using SAS Management Console must also be made available to the Data category view. For more information, see Creating a Performance Table.
  • When you run performance monitoring reports, you can set up performance index alert and warning conditions to notify users if conditions exceed the indexes. For more information, see Performance Index Warnings and Alerts.
Last updated: February 22, 2017