SAS Certified Predictive Modeler Using SAS Enterprise Miner 14

The SAS Certified Predictive Modeler Using SAS Enterprise Miner 14 exam expands your knowledge of data preparation and gives you a better understanding of how to analyse and deploy models. You’ll also learn how to create prediction models and do pattern analysis.

Exam Prerequisites:

This examination will definitely help you achieve heights in your career. However, before planning to give this examination you are required to have below mentioned skills:

• Candidates should have a firm understanding and mastery of the functionalities for predictive modeling available in SAS Enterprise Miner.

Exam Details: SAS Predictive Modeler Using SAS Enterprise Miner 14

• The SAS Certified Predictive Modeler Using SAS Enterprise Miner 14 examination consists of 55-60 multiple-choice and short-answer questions.
• To successfully pass this examination you will need to achieve a score of 725 between a score range from 200 to 1,000 points.
• The total time allotted to complete the examination is 165 minutes. The cost of this examination is $250.

Exam Registration

Planning to register for the examination, no worries we have covered all the steps and information required to register. Follow the steps mentioned below:

• Pearson VUE is the official exam partner for SAS. You are required to go to the official website of SAS Certified Predictive Modeler Using SAS Enterprise Miner 14.
• Click on register for the exam
• If you already have an account with SAS, then login to your account, otherwise, signup for a new account.
• Use exam ID A00-255; required when registering with Pearson VUE.
• Follow the prompt and complete your registration.
• You will receive a confirmation mail once you complete your registration
• You should keep all the login details safely, for future use.

Course Outline

1. Data Sources -20-25%

Create data sources from SAS tables in Enterprise Miner

• Use the Basic Metadata Advisor
• Use the Advanced Metadata Advisor
• Customize the Advanced Metadata Advisor
• Set Role and Level metadata for data source variables
• Set the Role of the table (raw, scoring, transactional, etc)
• SAS Documentation: Data Source

Explore and assess data sources

• Create and interpret plots, including Histograms, Pie charts, Scatter plot, Time series, Box plot (SAS Documentation: The Basics of Creating Graphs with SAS/GRAPH Software)

• Identify distributions
• Find outlying observations
• Name number (or percent) of missing observations
• Find levels of nominal variables
• Explore associations between variables using plots by highlighting and selecting data
• Compare balanced and actual response rates when oversampling has been performed
• Explore data with the STAT EXPLORER node. (SAS Documentation: StatExplore Node)
• Explore input variable sample statistics
• Browse data set observations (cases)

Modify source data

• Replace zero values with missing indicators using the REPLACEMENT node (SAS Documentation: Replacement Node)
• Use the TRANFORMATION node to be able to correct problems with input data sources, such as variable distribution or outliers. (SAS Documentation: Transform Variables Node)
• Use the IMPUTE node to impute missing values and create missing value indicators (SAS Documentation: Impute Node)
• Reduce the levels of a categorical variable
• Use the FILTER node to remove cases (SAS Documentation: Filter Node)

Prepare data to be submitted to a predictive model

• Select a portion of a data set using the SAMPLE node (SAS Documentation: SAMPLE node)
• Partition data with the PARTITION Node (SAS Documentation: Data Partition Node)
• Use the VARIABLE SELECTION node to identify important variables to be included in a predictive model. (SAS Documentation: Variable Selection Node)
• Make use of the PARTIAL LEAST SQUARES node to identify important variables to be included in a predictive model. (SAS Documentation: Partial Least Squares Node)
• Use a DECISION TREE or REGRESSION nodes to identify important variables to be included in a predictive mode (SAS Documentation: Decision Tree Node)

2. Building Predictive Models – 35-40%

Describe key predictive modeling terms and concepts

• Data partitioning: training, validation, test data sets (SAS Documentation: Partition the Data)
• Observations (cases), independent (input) variables, dependent (target) variables
• Measurement scales: Interval, ordinal, nominal (categorical), binary variables
• Prediction types: decisions, rankings, estimates
• Dimensionality, redundancy, irrelevancy
• Decision trees, neural networks, regression models (SAS Documentation: Decision Tree)
• Model optimization, overfitting, underfitting, model selection
• Describe ensemble models

Build predictive models using decision trees

• Explain how decision trees identify split points (SAS Documentation: The TREESPLIT Procedure)
• Build decision trees in interactive mode
• Change splitting rules (SAS Documentation: The HPSPLIT Procedure)
• Explain how missing values can be handled by decision trees
• Assess probability using a decision tree (SAS Documentation: Split best)
• Prune decision trees (SAS Documentation: Prune a Decision Tree)
• Adjust properties of the DECISION TREE node, including: subtree method, Number of Branches, Leaf Size, Significance Level, Surrogate Rules, Bonferroni Adjustment (SAS Documentation: Decision Tree Node)
• Interpret results of the decision tree node, including: trees, leaf statistics, treemaps, score rankings overlay, fit statistics, output, variable importance, subtree assessment plots
• Explore model output (exported) data sets (SAS Reference: Exporting a SAS data set in SAS format)

Build predictive models using regression

• Explain the relationship between target variable and regression technique
• Describe linear regression (SAS Documentation: Regression Model)
• Explain logistic regression (Logit link function, maximum likelihood) (SAS Reference: A Guide to Logistic Regression in SAS)
• Explain the impact of missing values on regression models (SAS Reference: Missing Data Analysis)
• Select inputs for regression models using forward, backward, stepwise selection techniques
• Adjust thresholds for including variables in a model
• Interpret a logistic regression model using log odds
• Interpret the results of a REGRESSION node (Output, Fit Statistics, Score Ranking Overlay charts) (SAS Documentation: Regression Node)
• Use fit statistics and iteration plots to select the optimum regression model for different decision types
• Add polynomial regression terms to regression models. (SAS Documentation: Polynomial Regression)
• Determine when to add polynomial terms to linear regression models

Build predictive models using neural networks

• Theory of neural networks (Hidden units, Tanh function, bias vs intercept, variable standardization)
• Build a neural network model (SAS Documentation: Neural Networks)
• Use regression models to select inputs for a neural network
• Explain how neural networks optimize their model (stopped training)
• Recognize overfit neural network models.
• Interpret the results of a NEURAL NETWORK node, including: Output, Fit Statistics, Iteration Plots, and Score Rankings Overlay charts (SAS Documentation: Neural Network Node)

3. Predictive Model Assessment and Implementation – 25-30%

Use the correct fit statistic for different prediction types

• Misclassification (SAS Documentation: Misclassification Matrix)
• Average Square Error
• Profit/Loss
• Other standard model fit statistics

Use decision processing to adjust for oversampling (separate sampling)

• Explain reasons for oversampling data (SAS Documentation: oversampling)
• Adjust prior probabilities
• Use profit/loss information to assess model performance
• Build a profit/loss matrix (SAS Documentation: Decision Thresholds and Profit Charts)
• Add a profit/loss matrix to a predictive model
• Determine an appropriate value to use for expected profit/loss for primary outcome (SAS Documentation: Decisions)
• Optimize models based on expected profit/loss

Compare models with the MODEL COMPARISON node

• Model assessment statistics (SAS Documentation: Model Fit and Assessment Statistics)
• ROC Chart (SAS Documentation: Receiver Operating Characteristic (ROC) Curves)
• Score Rankings Chart, including (cumulative) % response chart, (cumulative) Lift chart, gains chart.
• Total expected profit
• Effect of oversampling

Score data sets within Enterprise Miner

• Configure a data set to be scored in Enterprise Miner
• Use the SCORE node to score new data (SAS Documentation: Score Node)
• Save scored data to an external location with the SAVE DATA node (SAS Documentation: Save Data Node)
• Export SAS score code (SAS Documentation: Generating and Using Scoring Code)

4. Pattern Analysis – 10-15%

Identify clusters of similar data with the CLUSTER and SEGMENT PROFILE nodes

• Select variables to use to define the clusters (SAS Documentation: Cluster Analysis)
• Standardize variable scales
• Explore clusters with results output and plots
• Compare the distribution of variables within clusters

Perform association and sequence analysis (market basket analysis)

• Explain association concepts (Support, confidence, expected confidence, lift, the difference between association and sequence rules) (SAS Documentation: Strength of Association for Concept Linking)
• Create a data set for association analysis
• Interpret the results and graphs of the ASSOCIATION node. (SAS Documentation: Association Node)

For more information, click on SAS Certified Predictive Modeler Using SAS Enterprise Miner 14 FAQ.

Preparatory Guide for SAS Certified Predictive Modeler Using SAS Enterprise Miner 14

All set for your examination. Now it’s to time start your preparation. Here, we have provided you will all the necessary study material which you need to pass the examination.

1. Refer to the Exam Guide

When you are preparing for any examination it is important to know the exam objectives. For SAS Certified Predictive Modeler Using SAS Enterprise Miner 14 examination we have provided you with those exam objectives which are important from an examination point of view.

• Data Sources – 20-25%
• Building Predictive Models – 35-40%
• Predictive Model Assessment and Implementation – 25-30%
• Pattern Analysis – 10-15%  

2. Learning Resources

Now when you are clear with the exam objectives, it is time to start your preparation. For that you need study material, no worries, we are here to help you out.

Training:

SAS has provided official training which will help you develop the relevant skills required for the examination. The Applied Analytics Using SAS Enterprise Miner training teaches you how to create analytical flow diagrams using SAS Enterprise Miner’s powerful toolset for pattern identification (segmentation, association, and sequence analyses) and predictive modeling (decision tree, regression, and neural network models).

Reference Books:

Person’s best buddy is a book. You will need books to study for the exam if you want to pass it. Here are a few books to which you might refer when studying for the exam:

• Predictive Modeling With SAS® Enterprise Miner™: Practical Solutions for Business Applications by Kattamuri S. Sharma

3. Join Study Groups

Joining study groups is an excellent method to become totally immersed in the certification test for which you applied. These groups will assist you in keeping up to know with any recent modifications or exam updates. In addition, both novices and professionals are represented in these groups. You are free to ask any test-related question or discuss the exam without the fear of being judged. Furthermore, you may start a debate about any exam-related concern or query here. You will receive the best possible response to your inquiry if you do so.

4. Practice Test

It is very important to practice what you have learned so that you are in a position to analyze your practice, by practicing you will be able to improve your answering skills which will result in saving a lot of time. Moreover, the best way to start doing practice tests is after completing one full topic as this will work as a revision part for you. So, start practicing now!