Exam ID: A00-272
Exam Name: SAS Certified Visual Modeler Using SAS Visual Statistics 7.4
Successful candidates should be skilled in topics such as
Try Online Exam »
- Building and exploring descriptive models
- Building and exploring predictive models with continuous and categorical targets
- Assessing model goodness of fit
- Modifying and comparing models
- Scoring models.
SAS A00-272 Exam Summary:
Exam Name | SAS Certified Visual Modeler Using SAS Visual Statistics 7.4 |
Exam Code | A00-272 |
Exam Duration | 90 minutes |
Exam Questions | 50 Multiple choice questions |
Passing Score | 68% |
Exam Price | $180 (USD) |
Training & Books | SAS Visual Statistics: Interactive Model Building & An Introduction to SAS Visual Analytics |
Sample Questions | SAS Visual Modeler Certification Sample Question |
Practice Exam | SAS Visual Modeler Certification Practice Exam |
SAS A00-272 Exam Topics:
Objective | Details |
SAS® Visual Statistics Cross-functional Tasks – 18% | |
Prepare data using SAS® Visual Analytics. | – Manage explorations and visualizations. – Impute a variable. – Transform a variable. – Create an aggregated measure. – Replace dirty data with missing values. – Combine multiple categories into fewer levels. – Create dummy variables in SAS® Visual Analytics and SAS® Visual Data Builder. |
Filter data used for a model. | – Exclude selections to filter data. – Apply filters to visualization and data source. – Review Measure Details. |
Use interactive group-by. | – Explain group-by modeling. – Assign a group-by variable to a predictive model (logistic regression, linear regression model and generalized linear model). – Interactively examine the Fit Summary for group-by models. – Choose the best fitting group-by model using fit statistics and Variable Importance. – Interpret model results using the advanced group-by feature. – Examine the summary table for group-by processing. |
Building and Assessing Segmentation Models – 32% | |
Perform unsupervised segmentation using cluster analysis. | – Explain the unsupervised classification. – Given a scenario, set proper inputs for k-means algorithm. – Build a cluster analysis in SAS® Visual Statistics. – Assign roles for cluster analysis. – View and edit cluster properties. – Set Parallel Coordinate properties for a cluster. – Given a scenario, appropriately change the number of clusters. – Derive a cluster ID variable and use it in another visualization. |
Analyze cluster results. | – Interpret a Cluster Matrix. – Interpret Parallel Coordinates plot. – Interpret Cluster Summary tab. |
Perform supervised segmentation using decision trees. | – Explain how split points are determined. – Assign variable roles for a decision tree. – Define decision tree properties. – Describe how predictions are formulated for a decision tree. – Explain variable selection methods for decision trees. – Derive a leaf ID for use in other models. – Prune a decision tree. |
Asses decision tree results. | – Interpret tree with Tree Map. – Interpret Leaf statistics. – Interpret Assessment panel. – Investigate leaf nodes. – Explain icicle plot. |
Building and Assessing Regression-type Models – 40% | |
Explain linear models. | – Explain linear regression. – Model effects usage. – Given a scenario, determine when to use a linear regression model vs. a generalized linear model. |
Perform linear regression modeling. | – Assign linear regression roles. – Add Interaction Effect. – Define linear regression properties. – Explain informative missingness. – Review outlier details and exclude outliers. |
Perform generalized linear regression modeling. | – Assign generalized linear model roles. – Assign offset variable. – Define linear regression properties. – Link functions and distributions in generalized linear models. – Given a scenario, choose appropriate distribution and link function. |
Perform logistic regression modeling. | – Explain logistic regression essentials. – Explain prediction in logistic regression. – Explain variable selection in SAS® Visual Statistics. – Specify which variable is the event (binary). – Specify how a multinomial response variable is used in SAS® Visual Statistics. – Assign logistic regression roles. – Define logistic regression properties. – Specify when to use appropriate link function when building a predictive model. |
Assess model results. | – Interpret Fit Summary window. – Interpret Residual Plot. – Interpret ROC chart (KS Statistic). – Evaluate the Misclassification plot. – Evaluate the Lift chart. – Interpret Influence plot. – Interpret Summary bar. – Assess residuals and other model diagnostics to choose an appropriate distribution and link function. – Derive predicted values and describe in terms of predicted probabilities in SAS® Visual Statistics. – Apply prediction cut-off. |
Model Comparison and Scoring – 10% | |
Compare models | – Explain model comparison features. – Assign model comparison properties. – Interpret comparison results using Assessment panel, Fit Statistics, ROC charts, concordance statistics, misclassification, etc. – Interpret Summary Table for model comparison (statistics, variable importance). – Given a scenario, use a particular fit statistic to select a champion model. – Define the conditions that make models comparable in SAS® Visual Statistics. |
Score models | – Explain scoring functionality. – Export score code. – Implement score code. – Identify which SAS® tools can score new data using score code generated by SAS® Visual Statistics. |