SAS AI and Machine Learning practice questions
SAS · SAS-AIML · 300 questions
Original practice questions for SAS AI and Machine Learning.
This course contains the use of artificial intelligence.
Practice Quizzes
Test your knowledge with standard 20-question practice sets.
Quiz 1
Quiz 2
Quiz 3
Quiz 4
Quiz 5
Quiz 6
Quiz 7
Quiz 8
Quiz 9
Quiz 10
Quiz 11
Quiz 12
Quiz 13
Quiz 14
Quiz 15
Browse by Domain
Study specific topics at your own pace.
Domain 1: Machine Learning Fundamentals · 47 questions
- A data scientist is training a gradient boosting model in SAS and notices the training error continues to decrease while validation error increases after a certain number of iterations. Which concept best describes this phenomenon?
- When performing k-fold cross-validation in SAS Model Studio, what is the primary purpose of holding out each fold in turn?
- A model trained on a small dataset achieves 99% training accuracy but only 62% test accuracy. Which remediation strategy would most directly address the bias-variance tradeoff issue present?
- In SAS, a confusion matrix shows: TP=80, FP=20, FN=10, TN=90. What is the model's precision?
- A data scientist needs to compare two classifiers when the class distribution is highly imbalanced (95% negative, 5% positive). Which metric is most appropriate?
- Which cross-validation strategy is most appropriate when the dataset contains a strong time dependency, such as stock price data?
- The AIC (Akaike Information Criterion) penalizes model complexity. Which of the following correctly describes the AIC formula used in SAS model selection?
- When constructing a train/test split in SAS, a data scientist sets the random seed before calling PROC SURVEYSELECT. What is the main purpose of setting the seed?
- A data scientist observes that a model performs well on training data but poorly on new data. They decide to apply L2 regularization. In the context of the bias-variance tradeoff, what effect does increasing the L2 penalty (lambda) have?
- A data scientist constructs a ROC curve for a binary classifier. The curve passes exactly through the point (0.0, 1.0). What does this indicate about the model?
- In stratified k-fold cross-validation for a classification problem, what makes it 'stratified'?
- A model selection process uses BIC (Bayesian Information Criterion) instead of AIC. What is the key difference between BIC and AIC in model selection?
- A regression model produces residuals that exhibit a funnel shape when plotted against fitted values. Which diagnostic does this indicate and what remediation is recommended?
- In a multi-class classification problem with 5 classes, a data scientist in SAS wants to use a One-vs-Rest (OVR) strategy with PROC HPLOGISTIC. How many binary models must be trained?
- When computing the F1 score for an imbalanced binary classification problem, which formula is correct?
- A data scientist observes that a gradient boosted model in SAS Viya has very low training error but high validation error, while a linear model has moderate error on both. The team wants to improve the gradient boosting model without collecting more data. Which technique is specifically designed to address this situation?
- A data scientist constructs a confusion matrix and finds the following values: TP=45, FP=5, FN=15, TN=135. What is the recall (sensitivity) of the model?
- A data scientist must choose between a model with AUC=0.82 and another with AUC=0.79. However, the business requires maximizing recall at a minimum precision of 0.70. Which approach best addresses this requirement?
- A data scientist is selecting between a model with 10 parameters (AIC=245) and one with 25 parameters (AIC=238). Which model should be selected and why?
- A model's learning curve shows that training error and validation error have both converged to a high value (approximately 0.35 RMSE) and neither decreases as more training data is added. What does this indicate?
- A data scientist uses PROC SURVEYSELECT in SAS to create a stratified random sample for model building, stratifying by a 'region' variable. Which option ensures proportional allocation across strata?
- In a binary classification problem, a data scientist reports that a model achieves 'lift of 3.5 at the first decile.' What does this mean?
- A SAS data scientist needs to compare the performance of PROC HPFOREST against PROC HPLOGISTIC using 5-fold cross-validation on the same data. Which approach correctly implements this comparison in SAS?
- A data scientist in SAS Viya wants to apply oversampling to a minority class before training PROC HPFOREST. Which SAS procedure generates a synthetic oversampled training dataset using SMOTE (Synthetic Minority Oversampling Technique)?
- A data scientist observes that increasing training set size improves validation accuracy but training accuracy stays fixed near 100%. This learning curve pattern is characteristic of which condition?
- Which regularization technique simultaneously performs variable selection by producing exactly zero coefficients for irrelevant features?
- A model achieves AUC = 0.50 on the test set. What does this indicate?
- When using k-fold cross-validation with k=5, how many times is each observation used for validation?
- A classification model on a 95% negative / 5% positive dataset predicts 'negative' for every observation. What is the F1 score for the positive class?
- In the context of model evaluation, what does 'calibration' refer to?
- A machine learning practitioner uses 5x2 cross-validation to compare two models. Which statistical test is specifically designed for this comparison procedure?
- A random forest model with 500 trees shows no improvement in OOB error beyond 200 trees. What does this demonstrate about random forests?
- A data scientist uses the Matthews Correlation Coefficient (MCC) instead of accuracy on an extremely imbalanced dataset where 98% of records are class 0. What is the key advantage of MCC in this situation?
- A model reliability diagram (calibration plot) shows that predicted probabilities near 0.8 correspond to observed event rates of only 0.6. What does this indicate about the model?
- Which cross-validation approach is specifically recommended for time series data where the relationship between features and target evolves over time?
- A team encodes the categorical variable 'state' (50 categories) by replacing each state with the average default rate computed on training data. What encoding method is this, and what is its primary risk?
- A data scientist uses Platt scaling to calibrate a gradient boosting model's output. What does Platt scaling specifically fit to the raw model scores?
- For a logistic regression credit risk model, Weight of Evidence (WOE) encoding is applied to a high-cardinality categorical variable. Why is WOE particularly suited for logistic regression?
- Log-loss penalizes confident wrong predictions much more heavily than uncertain wrong predictions. Which scenario produces the highest log-loss for a single observation?
- A practitioner computes Cohen's Kappa = 0.72 for a 3-class text classifier. What does this value indicate according to the Landis and Koch scale?
- A data scientist needs model-agnostic local explanations for individual predictions from a gradient boosting model. Which technique provides additive feature attributions that sum to the difference between the model's output and the global expected output?
- A credit risk team builds a logistic regression model and reports Gini = 0.62. How is the Gini coefficient derived from the AUC?
- A machine learning team trains a neural network and observes that training loss decreases smoothly but validation loss exhibits high variance across epochs. Which regularization technique specifically reduces variance by injecting Gaussian noise into the input layer during training?
- A researcher applies SMOTE (Synthetic Minority Over-sampling Technique) to address class imbalance before training. What synthetic minority samples does SMOTE generate?
- Information Criterion (AIC and BIC) are used to compare competing regression models. Which statement correctly differentiates AIC from BIC for model selection?
- A practitioner builds a stacking ensemble with logistic regression, gradient boosting, and random forest as base learners, and a ridge regression as the meta-learner. What is the critical data hygiene requirement for training the meta-learner?
- A machine learning team applies partial dependence plots (PDPs) to interpret a random forest model's relationship between 'customertenure' and predicted churn probability. The PDP shows a non-monotone U-shaped curve. What does this indicate?
Domain 2: SAS Viya & SAS Studio · 45 questions
- In SAS Viya, a data engineer wants to load a large CSV file directly into a CAS in-memory table for model training. Which CAS action set is most appropriate for this task?
- A SAS programmer needs to pass a dynamic list of variables to a macro that builds a model. Which SAS macro technique allows iterating over a space-delimited list of variable names?
- A data scientist uses PROC FEDSQL in SAS Viya to query data across heterogeneous sources. What is the primary advantage of PROC FEDSQL over PROC SQL in a Viya environment?
- In a SAS DATA step, a programmer writes: IF revenue > 100000 THEN tier = 'Gold'; ELSE IF revenue > 50000 THEN tier = 'Silver'; ELSE tier = 'Bronze'; Which SAS behavior occurs when revenue is a missing value (.)?
- A SAS administrator wants to use PROC SQL to identify duplicate customer records based on email address. Which approach correctly identifies duplicate email addresses?
- When working in SAS Studio, a user creates a flow in the Visual Analytics workflow. Which component of SAS Viya manages the distributed in-memory execution of that flow?
- A SAS programmer discovers that two tables need to be joined on a common key but the key variable has different names: 'custid' in Table A and 'customerid' in Table B. Which PROC SQL join syntax correctly handles this?
- In SAS Viya, a data scientist wants to list all CAS tables available in a specific caslib called 'MYDATA'. Which CAS action accomplishes this?
- A SAS data scientist needs to run PROC HPFOREST inside a CAS session. Which step must be completed before executing the procedure to ensure it runs in distributed mode?
- A SAS programmer uses the %SYSFUNC macro function in the following code: %let today = %sysfunc(today(), date9.); What does this code accomplish?
- In SAS Viya, a data engineer uses the following CAS action call: cas.table.save(table={name='MYMODEL', caslib='MODELS'}, name='mymodel.sashdat', caslib='MODELS'); What does this accomplish?
- A SAS programmer writes: DATA output; SET input; IF N = 1 THEN DO; ... END; RUN; What is the purpose of the IF N = 1 condition?
- A SAS data scientist analyzes a CAS dataset using the correlation action. The following code is submitted: proc cas; simple.correlation / table={name='CUSTOMERS', caslib='PUBLIC'} inputs={'age','income','spend'} / casout={name='corrout', caslib='WORK'}; quit; What is the output stored in 'corrout'?
- A SAS programmer writes a macro to generate model code. The macro uses %LET inside the macro definition. What is the scope of macro variables created with %LET inside a macro?
- In SAS Viya, a data scientist wants to use the automl CAS action to automatically search for the best model for a classification target. Which action set contains the automl action?
- A SAS programmer is building a scoring pipeline in DS2. Which statement correctly defines a DS2 package method signature for a scoring function that takes two numeric inputs and returns a numeric prediction?
- A SAS data scientist uses PROC HPREG with a SELECTION statement to perform variable selection. After the selection, the analyst wants to examine diagnostics for influential observations. Which PROC step should be run after PROC HPREG to examine Cook's Distance for the final selected model?
- A SAS programmer uses the following code to create a macro variable from a dataset count: PROC SQL; SELECT COUNT() INTO :nobs TRIMMED FROM mydata; QUIT; What does TRIMMED accomplish in this context?
- In SAS Viya, a data scientist submits a CAS action using Python's SWAT package: import swat; s = swat.CAS('myserver', 5570); s.loadactionset('decisionTree'); What does this code do?
- A data engineer uses PROC CASUTIL to copy a table from one caslib to another in SAS Viya. Which statement correctly copies table 'SALES' from caslib 'WORK' to caslib 'MYLIB'?
- A SAS Viya analyst runs PROC FEDSQL to query a table stored in a Hadoop Hive database registered as a caslib. Which type of processing does PROC FEDSQL enable for this query?
- A data scientist uses the following SAS code: DATA scored; SET newcustomers; ARRAY wts{5} w1-w5; pred = 0; DO i = 1 TO 5; pred + wts{i} input{i}; END; DROP i; RUN; What is this DATA step implementing?
- In SAS Viya, which statement correctly establishes a connection to a CAS server within a SAS program?
- A SAS programmer needs to execute a CAS action to list all tables currently loaded in the CASUSER caslib. Which PROC CAS invocation is correct?
- In a SAS DATA step, which approach correctly reads the 3rd column (columns 15-16) as a 2-character state code from a fixed-width text file?
- What is the primary purpose of the %SYSFUNC function in SAS macro language?
- When PROC SQL references tables in a CAS library via a CAS libref in SAS Viya, where does SQL execution typically occur?
- In SAS Studio, which file type allows developers to create reusable programs with a graphical form interface for end users without requiring them to write code?
- What is the primary effect of the RETAIN statement in a SAS DATA step?
- PROC CASUTIL is used to load a CSV file from disk into a CAS in-memory table. Which statement correctly loads a CSV named 'customers.csv' from the caslib path into a CAS table named 'CUSTOMERS'?
- In SAS Viya, which capability allows external Python scikit-learn models to be registered in SAS Model Manager for centralized governance?
- In SAS Viya, what distinguishes the SAS Job Execution Service from the SAS Compute Service?
- A SAS developer uses PROC OPTMODEL to solve a logistics optimization problem. What type of optimization does PROC OPTMODEL natively support?
- In SAS Intelligent Decisioning, a Branch node evaluates a credit score against thresholds. What is the primary advantage of using Business Rule nodes over embedding IF-THEN logic in a SAS DATA step for this routing?
- A SAS developer automates CI/CD for model retraining by using the SAS Viya REST API to submit a job when new data arrives. Which SAS Viya service endpoint is used to submit a SAS program as a batch job?
- SAS Model Studio's Model Comparison view displays results for logistic regression, gradient boosting, and random forest models after an AutoML pipeline run. Which metric does SAS Model Studio use by default to rank classification models in this comparison?
- A SAS developer needs to join a SAS dataset in the SAS work library with a CAS in-memory table. Which approach correctly accomplishes this in SAS Viya?
- A SAS developer builds a macro %LOADDATA that accepts a table name and caslib. The macro should execute different code paths depending on whether the CAS table already exists. Which macro function checks for CAS table existence before loading?
- SAS DS2 is used for high-performance scoring in CAS instead of the traditional DATA step. What is the primary advantage of PROC DS2 over a DATA step in the CAS context?
- A SAS developer writes a macro that calls %SYSFUNC(CATS(prefix,&suffix)) to build a variable name dynamically. What does the CATS function do in this context?
- PROC FEDSQL is used to query across a CAS in-memory table and an external RDBMS table simultaneously. Which characteristic distinguishes PROC FEDSQL from PROC SQL in this federated scenario?
- A SAS Viya CAS administrator needs to promote an in-memory CAS table so it persists across CAS sessions and is visible to all users. Which PROC CASUTIL operation accomplishes this?
- A SAS developer uses %MACRO with the MINDELIMITER= and PARMBUFF options. What does the PARMBUFF option enable?
- A SAS/ACCESS connection uses PROC SQL with a pass-through query to execute native Oracle SQL. Which statement correctly implements an explicit SQL pass-through query to Oracle?
- In SAS Viya, CAS action sets provide the interface to CAS analytics capabilities. A developer calls cas.simple.summary() in Python via swat. What is a CAS action set, and how does it differ from a SAS procedure?
Domain 3: Supervised Learning with SAS · 61 questions
- A data scientist runs PROC HPLOGISTIC on a binary outcome in SAS. Which statement correctly sets the reference level for a CLASS variable named 'Region' to 'West'?
- A data scientist trains a random forest model using PROC HPFOREST in SAS and wants to limit the maximum number of leaf nodes in each tree to control overfitting. Which option accomplishes this?
- In PROC HPNEURAL, which activation function is specified for hidden layer neurons using the HIDDEN statement?
- A data scientist uses PROC HPREG in SAS to perform LASSO variable selection. Which option enables LASSO regularization?
- When using gradient boosting in SAS Viya's gbtreetrain CAS action, what does the SUBSAMPLERATE= parameter control?
- A decision tree trained with PROC HPSPLIT in SAS has grown too deep and is overfitting. Which pruning approach is available in PROC HPSPLIT?
- In PROC HPLOGISTIC, which fit statistic should a data scientist examine to evaluate whether adding a predictor significantly improves the model compared to a nested simpler model?
- A data scientist runs PROC HPFOREST with SEED=12345 and obtains variable importance scores. The top variable is 'creditscore'. Which metric does PROC HPFOREST use by default to compute variable importance?
- Which SAS procedure is purpose-built for training gradient boosting decision trees (GBDT) in the SAS 9.4 environment (not Viya)?
- In a SAS Viya pipeline, PROC HPLOGISTIC is used to build a binary classification model. The analyst wants to obtain predicted probabilities for the positive class. Which output statement is correct?
- In PROC HPLOGISTIC output, the 'Convergence Status' indicates 'Convergence criterion (GCONV=1E-8) satisfied.' What does this mean?
- A data scientist applies PROC HPFOREST to a regression problem. Which option in PROC HPFOREST specifies the number of variables randomly considered at each split?
- In SAS, a data scientist wants to use cost-sensitive learning to handle a class imbalance problem in PROC HPLOGISTIC. Which approach correctly applies differential misclassification costs?
- A data scientist trains a neural network using PROC HPNEURAL with 2 hidden layers. Which option in the PROC statement controls the maximum number of training iterations?
- When using PROC HPREG with forward stepwise selection, which criterion controls which variables are added to the model at each step?
- A SAS data scientist uses the following code: PROC HPFOREST DATA=train SEED=999; TARGET churn / LEVEL=NOMINAL; INPUT age income tenure / LEVEL=INTERVAL; INPUT segment / LEVEL=NOMINAL; OOB; RUN; What does the OOB statement request?
- In PROC HPSPLIT for classification trees, which impurity measure is the default splitting criterion?
- A data scientist uses PROC HPREG to fit a polynomial regression with terms up to degree 3. How should the cubic term for variable 'X' be specified in the MODEL statement?
- A SAS data scientist submits the following PROC HPNEURAL code and gets a 'CONVERGENCE NOT ACHIEVED' warning. Which adjustment is most likely to resolve convergence issues?
- A data scientist uses PROC HPLOGISTIC and requests the ROC curve output. Which ODS output object contains the ROC curve data points (sensitivity and 1-specificity pairs)?
- A data scientist in SAS Viya runs PROC HPREG and gets a Type I SS (Sequential Sum of Squares) and Type II SS (Partial Sum of Squares) table. When do the Type I and Type II SS values differ for a predictor?
- When using the SAS Viya decisionTree.dtreeTrain CAS action, which parameter controls the minimum number of training observations required before a node will be split?
- In SAS, which procedure is most appropriate for building a multivariate regression model when the number of predictors (p=500) greatly exceeds the number of observations (n=200)?
- A data scientist uses PROC HPNEURAL in SAS with TRAIN TECH=LEVMAR. What optimization algorithm does LEVMAR specify?
- In PROC HPLOGISTIC, a data scientist specifies LINK=PROBIT in the MODEL statement instead of the default LINK=LOGIT. What change does this make to the model?
- A data scientist trains a PROC HPFOREST model on a credit scoring dataset. The training AUC is 0.95 but validation AUC is 0.71. Which PROC HPFOREST option is most likely to close this gap without changing the dataset?
- In PROC HPLOGISTIC, which option generates a score dataset with predicted probabilities for each new observation?
- A data scientist uses the gradient boosting gbtreetrain CAS action in SAS Viya and wants to tune the learning rate (shrinkage) parameter. What is the effect of decreasing the learning rate from 0.1 to 0.01 while keeping all other parameters constant?
- A data scientist specifies PROC HPSPLIT with a GROW ENTROPY option. What splitting criterion does this select?
- A data scientist runs PROC HPNEURAL for regression. Which option correctly specifies that the output layer should use a linear (identity) activation function for a continuous target?
- In a SAS Viya gradient boosting model (gbtreetrain CAS action), a data scientist sets SUBSAMPLERATE=0.8 and COLSAMPLEBYTREE=0.7. What do these parameters together create?
- A data scientist trains a decision tree in PROC HPSPLIT. The output shows a leaf node with 3 observations, all of the same class. What is the Gini impurity of this node?
- A SAS data scientist is preparing features for a neural network in PROC HPNEURAL. Input variables have widely different scales: 'age' ranges 18-90, 'income' ranges 20000-500000, and 'loanamount' ranges 1000-50000. Why is standardization critical before training?
- A data scientist uses PROC HPREG with SELECTION=STEPWISE and SLENTRY=0.05 SLSTAY=0.10. What do these two significance level options control?
- In SAS Viya's gradient boosting implementation, a data scientist wants to prevent individual trees from memorizing specific training observations. Which combination of parameters directly implements this goal?
- For a binary classification problem, which splitting criterion does PROC HPFOREST use by default?
- In PROC HPREG diagnostics, a predictor has VIF = 18. What does this indicate and what is the typical remediation?
- In PROC LOGISTIC with CLASS variables, which parameterization uses the last alphabetically sorted category as the reference by default?
- A gradient boosting model with 1000 trees and learning rate 0.01 shows overfitting. Which regularization approach is most effective without reducing tree count?
- In PROC HPNEURAL, which statement and option specify the number of neurons in a hidden layer?
- Why is PROC SURVEYLOGISTIC preferred over PROC LOGISTIC for survey data analysis?
- How does PROC HPFOREST handle missing values in predictor variables during training and scoring?
- PROC HPFOREST is trained with DEPTH=10 on a dataset with 100 predictor variables. A colleague suggests reducing DEPTH=4. What is the primary effect of reducing tree depth in a random forest?
- PROC HPREG is run with SELECTION=STEPWISE and the default entry/removal significance levels. A variable 'creditscore' is added at step 2 but then removed at step 6 when 'paymenthistory' enters. At step 9, 'creditscore' re-enters after 'debtratio' removes 'paymenthistory'. What is the most appropriate modeling response to this stepwise cycling?
- PROC HPFOREST generates an OOB error estimate as part of its default output. What distinguishes the OOB error estimate from a standard holdout validation error?
- PROC GAMPL is used to model a non-linear relationship between patient age and hospital readmission. Which aspect of PROC GAMPL's penalized spline distinguishes it from including polynomial terms in PROC LOGISTIC?
- PROC LIFETEST is run with the STRATA= statement to compare survival curves between two treatment arms. By default, which test does PROC LIFETEST report?
- In PROC PHREG, a hazard ratio of 2.5 is reported for a binary predictor 'treatment' (0=control, 1=treated). How is this interpreted?
- A PROC HPNEURAL model is trained with MLPARCH(20 10) and the VALIDATION= option. Training loss decreases monotonically but validation loss begins rising at epoch 15. What does PROC HPNEURAL do when the VALIDATION= option is specified?
- In PROC GLMSELECT with SELECTION=LASSO, the PATH= option is specified. What does the solution path represent?
- A data scientist builds a PROC HPFOREST model and a PROC TREEBOOST model on the same data and compares their variable importance outputs. A colleague claims both should rank variables identically. Why is this assumption incorrect?
- A gradient boosting model is trained with SHRINKAGE=0.01 (low learning rate) and NTREES=500. A colleague trains the same model with SHRINKAGE=0.1 and NTREES=50. Assuming the same training dataset, what is the expected difference in model behavior?
- PROC HPREG is run on a dataset with 50 predictors. VIF values for three predictors are 22, 18, and 31. Which action is most appropriate based on these VIF values?
- PROC HPLOGISTIC is run for a binary fraud classification. The output shows concordance = 0.83, discordance = 0.14, and ties = 0.03. How is the c-statistic (AUC) computed from these values?
- A SAS analyst fits PROC GENMOD to count data representing hospital readmissions per patient. The output shows significant overdispersion (deviance/DF = 3.2). Which distribution family should replace Poisson to better handle overdispersion?
- PROC GLMSELECT with SELECTION=STEPWISE is run. A variable is added at step 3, removed at step 5, and added again at step 8. What does this cycling indicate?
- A SAS analyst uses PROC SURVEYSELECT to create a stratified random sample. Why is stratified sampling preferred over simple random sampling when the population contains small but important subgroups?
- PROC DISCRIM is used to classify observations into predefined groups. Which assumption does Linear Discriminant Analysis (LDA) in PROC DISCRIM make that Quadratic Discriminant Analysis (QDA) does not?
- A PROC HPFOREST model is trained with VARSTOTRY=5 on a dataset with 40 predictor variables. What does VARSTOTRY=5 control?
- A SAS developer builds a gradient boosting model using PROC TREEBOOST with the NTREES=200, SHRINKAGE=0.05, and SUBSAMPLE=0.8 options. What does SUBSAMPLE=0.8 control in stochastic gradient boosting?
- PROC LIFEREG is used to fit a Weibull accelerated failure time (AFT) model. How does PROC LIFEREG differ from PROC PHREG in its modeling approach?
Domain 4: Unsupervised Learning with SAS · 46 questions
- A data scientist runs PROC HPCLUS in SAS to perform k-means clustering on customer data. Which option specifies the maximum number of iterations for the algorithm to converge?
- When comparing k-means clustering and hierarchical clustering for a dataset with 500,000 observations in SAS, which statement is most accurate?
- A data scientist wants to reduce the dimensionality of a dataset with 50 numeric variables before clustering. In SAS, which procedure is most commonly used to perform Principal Component Analysis?
- In PROC HPCLUS, a data scientist wants SAS to automatically determine the optimal number of clusters. Which option enables this capability?
- A data scientist uses PROC HPDS2 in SAS Viya to score new data against an anomaly detection model. What is the primary role of PROC HPDS2 in this workflow?
- A data scientist applies the elbow method to choose k for PROC HPCLUS. The within-cluster sum of squares (WCSS) values are: k=2: 1200, k=3: 700, k=4: 500, k=5: 480, k=6: 465. Which k value does the elbow method suggest?
- A data scientist is evaluating whether to use PROC HPCLUS or PROC CLUSTER for a dataset with 2 million rows. Which statement best justifies the choice of PROC HPCLUS?
- A data scientist applies PCA via PROC PRINCOMP and retains components that explain at least 80% of total variance. The cumulative proportion output shows components 1-4 explain 82% of variance. How many components should be retained?
- In PROC HPCLUS, which initialization method assigns initial cluster seeds by selecting observations that are maximally separated from each other?
- A data scientist applies hierarchical clustering using PROC CLUSTER with METHOD=WARD. What does Ward's linkage criterion minimize at each merge step?
- A data scientist runs PROC FACTOR with METHOD=PRINCIPAL ROTATE=VARIMAX on a 20-variable dataset. What is the purpose of ROTATE=VARIMAX?
- In SAS, a data scientist applies PROC HPCLUS to segment customers and wants to profile each cluster by computing mean values of key variables. Which post-clustering approach accomplishes this most efficiently?
- A data engineer uses the t-SNE algorithm output from SAS to visualize high-dimensional customer segments. What is a key limitation of t-SNE that the engineer should communicate to stakeholders?
- A SAS analyst uses PROC HPCLUS to cluster 1 million customers. After obtaining cluster assignments, they want to score new customers arriving each week using the cluster centroids. Which approach is most appropriate?
- A data scientist applies PROC FACTOR with NFACTORS=4 and uses the SCREE plot output to decide the final number of factors. The eigenvalues are: Factor 1: 5.2, Factor 2: 3.1, Factor 3: 1.8, Factor 4: 1.1, Factor 5: 0.4. Using the Kaiser criterion (eigenvalue > 1), how many factors should be retained?
- A data scientist applies PROC HPCLUS and obtains cluster centroids. They want to interpret which input variables drive the most separation between clusters. Which approach best identifies the most discriminating variables?
- A data scientist applies PROC HPCLUS to 800,000 observations with k=8 clusters and converges in 15 iterations. The data scientist then wants to understand within-cluster variability. Which statistic output by PROC HPCLUS directly measures within-cluster variability?
- A data scientist uses PROC PRINCOMP in SAS with OUTSTAT= to save the component structure. Later, they want to score new data by projecting it onto the saved principal components. Which step applies the saved PCA transformation to new data?
- In a SAS k-means clustering workflow using PROC HPCLUS, the pseudo-F statistic (Calinski-Harabasz index) is computed across k=2 through k=10. The pseudo-F values are: k=3: 482, k=4: 716, k=5: 698, k=6: 534. Which k value does the pseudo-F criterion suggest?
- A data scientist runs PROC CLUSTER with METHOD=AVERAGE (unweighted average linkage) and then uses PROC TREE to create cluster assignments. What does average linkage use to compute the distance between two clusters?
- A data scientist builds an anomaly detection pipeline in SAS Viya. They want to flag observations whose Mahalanobis distance from the cluster center exceeds a threshold. Which SAS procedure or action set is most appropriate for computing Mahalanobis distances?
- A data scientist applies PROC HPCLUS with k=5 to customer transaction data. After initial cluster assignment, the data scientist suspects cluster 3 contains outliers. Which SAS approach best identifies outlier observations within a cluster?
- A data scientist runs PROC FACTOR and examines the communality estimates in the output. What does a communality value of 0.25 for a variable named 'age' indicate?
- After running PROC HPCLUS with NCLUST=4, two clusters appear very similar. Which metric best quantifies the separation between these two specific clusters?
- PROC FACTOR is run with ROTATE=VARIMAX. What is the purpose of Varimax rotation?
- Which PROC PRINCOMP option saves the principal component scores for each observation to a SAS dataset?
- PROC STDIZE is run with METHOD=MIDRANGE before clustering. What transformation does this apply?
- In hierarchical clustering with PROC CLUSTER METHOD=COMPLETE, how is the between-cluster distance defined?
- What advantage does PROC FASTCLUS offer for very large datasets compared to PROC HPCLUS?
- In isolation forests for anomaly detection, why are anomalies isolated in fewer random splits than normal points?
- A SAS PROC VARCLUS output shows that a cluster has R-squared with its own component = 0.45 but next-nearest cluster R-squared = 0.48. What does this imply?
- A researcher uses Gower distance for clustering a mixed dataset. Which PROC DISTANCE option computes Gower dissimilarity in SAS?
- The gap statistic is used to select the optimal number of clusters k. What is the reference distribution against which the observed log(WSS) is compared?
- How does fuzzy C-means clustering differ from k-means in its representation of cluster membership?
- After running PROC CLUSTER METHOD=WARD, a dendrogram is produced. Which PROC TREE option cuts the dendrogram at a specific number of clusters?
- A SAS analyst runs PROC PRINCOMP and requests the first 3 principal components. The cumulative variance explained is 45%, 62%, and 74%. Which statement about the 3rd principal component is correct?
- PROC CLUSTER with METHOD=AVERAGE (average linkage) computes the distance between two clusters being merged. How is this distance defined under average linkage?
- PROC DISTANCE with METHOD=MAHALANOBIS requires an additional option to compute Mahalanobis distance. What does Mahalanobis distance account for that Euclidean distance does not?
- PROC FASTCLUS is run on a 5-million-row dataset with MAXCLUSTERS=8 and MAXITER=20. Which characteristic makes PROC FASTCLUS more efficient than PROC HPCLUS for very large datasets when a fixed number of clusters is required?
- PROC ACECLUS is run as a preprocessing step before PROC CLUSTER. What does PROC ACECLUS estimate that improves subsequent clustering?
- PROC MODECLUS is applied to a multivariate dataset. What type of clustering algorithm does PROC MODECLUS implement?
- A data scientist uses PROC DISTANCE with METHOD=EUCLID on a dataset where predictors have very different scales (e.g., age: 18-80, income: 20000-500000). What problem does this create for clustering?
- PROC HPCLUS is run with NCLUST=5 on a 10-million-row dataset. The IMPUTE option is specified. What does IMPUTE do in PROC HPCLUS?
- A data scientist runs PROC PRINCOMP and uses the OUTSTAT= option. What information does the OUTSTAT= dataset contain?
- An isolation forest is trained to detect anomalies in a network traffic dataset. An observation receives an anomaly score close to 1.0. What does this score indicate?
- A practitioner uses PROC STDIZE with METHOD=RANGE to standardize a variable with min=10 and max=90. After standardization, a raw value of 50 maps to what standardized value?
Domain 5: Model Deployment & Monitoring · 49 questions
- In SAS Model Manager, what is the purpose of designating a model as the 'champion' model in a champion/challenger deployment pattern?
- A data scientist uses PROC ASSESS in SAS to evaluate a deployed logistic regression model. Which capability does PROC ASSESS provide that directly addresses model drift detection?
- In SAS Model Manager, a data scientist registers a SAS score code file. Which file extension does SAS Model Manager require for SAS score code?
- A MLOps engineer configures model monitoring in SAS Model Manager to detect data drift. Which statistical measure is commonly used to compare the distribution of an input variable between training data and current production data?
- A data scientist wants to retrieve a specific version of a registered model from SAS Model Manager via the SAS Viya REST API. Which resource path is correct?
- During model monitoring in SAS, a data scientist observes that the Gini coefficient for a credit risk model has declined from 0.62 at deployment to 0.41 six months later. What does this indicate?
- A model built in SAS Model Studio has been registered in SAS Model Manager. A data engineer wants to generate batch scores for 10 million new records overnight. Which SAS component is most appropriate for orchestrating this batch scoring job at scale?
- In SAS Model Manager, which term describes the automated process of detecting when a model's predictive performance has degraded enough to trigger an alert or retraining workflow?
- A data scientist needs to load a Python-trained scikit-learn model into SAS Model Manager for deployment alongside SAS models. Which SAS capability enables this?
- In SAS Model Manager, a data scientist wants to publish a model to a SAS Micro Analytic Service (MAS) destination for real-time scoring. Which statement best describes MAS?
- A model deployed in SAS Model Manager is monitored using PROC ASSESS. The data scientist wants to track Kolmogorov-Smirnov (KS) statistic over monthly scoring runs. What does a significantly increasing KS statistic over time indicate?
- In a SAS Viya environment, a data scientist wants to automate monthly model retraining when data drift is detected. Which SAS component is most appropriate for orchestrating this automated workflow?
- A model registry in SAS Model Manager contains version 1 (champion) and version 2 (challenger) of a credit risk model. Version 2 shows a Gini coefficient of 0.71 vs. 0.64 for version 1 in the challenger test. What action should the data scientist recommend?
- In SAS Viya's Model Manager, what is the purpose of the 'model comparison report' generated during champion/challenger evaluation?
- In SAS Model Studio, a data scientist selects the 'Gradient Boosting' node and wants to tune the number of trees using AutoTune. Which SAS Viya feature enables automatic hyperparameter search for gradient boosting?
- In SAS Model Manager, when a data scientist publishes a model to the 'SAS Micro Analytic Service' destination, which score code format is required?
- A model monitoring system in SAS detects that input variable 'creditutilization' has a PSI (Population Stability Index) of 0.31 over the past month. How should the data scientist interpret this?
- A data science team uses SAS Model Manager to track multiple model versions. Which version control capability does SAS Model Manager provide?
- A deployed model in SAS Viya shows concept drift — the relationship between features and the target has changed but the input distributions remain stable. Which monitoring approach best detects concept drift (as opposed to data drift)?
- In SAS Model Manager, a data scientist configures a performance monitoring job that compares model predictions to actual outcomes. Which SAS procedure is most commonly used within this monitoring workflow to generate model fit statistics?
- A data scientist registers a SAS model in the Model Manager repository and must specify the model's input and output variable definitions. Which input variable property must be correctly specified to ensure downstream scoring pipelines work correctly?
- A data scientist uses SAS Model Studio to build an automated machine learning (AutoML) pipeline. After model comparison, the best model is a gradient boosting model. Which SAS Model Studio action makes the selected model available for production scoring in SAS Model Manager?
- In SAS Model Manager, a data scientist runs a monitoring report and observes that the model's KS statistic dropped from 0.55 (baseline) to 0.32 (current month). The input variable PSI values are all below 0.10. What type of drift does this pattern most likely indicate?
- A SAS Model Manager project has a model registered with only SAS score code (no ASTORE). How can this model be deployed for scoring?
- Which SAS Viya service provides REST API endpoints for submitting jobs, monitoring status, and retrieving results for integration with external orchestration tools?
- A monitoring threshold triggers PSI > 0.20 as an alert. What type of drift does PSI specifically measure?
- In SAS Model Manager, what does the formal 'model validation' workflow step accomplish before deployment?
- PROC ASSESS with the FITPLOT option is run after logistic regression. What does the FITPLOT produce?
- A model governance team wants to ensure a model does not produce significantly different false positive rates across demographic groups before deployment. Where in SAS Model Manager governance is this addressed?
- When a champion model is replaced by a challenger in SAS Model Manager, what happens to the previous champion?
- A data scientist registers a model in SAS Model Manager and wants to test it using SAS Micro Analytic Service (MAS) before champion promotion. What endpoint pattern does MAS expose for real-time score code testing?
- In SAS Model Manager, the champion/challenger framework is configured for a credit scoring model. The challenger receives 20% of scoring requests. After 30 days, the challenger's Gini (0.71) exceeds the champion's Gini (0.66) with statistical significance. What action does the champion/challenger framework enable?
- In SAS Model Manager, a model monitoring job runs weekly and computes PSI and KS statistics. The monitoring job configuration requires specifying a 'baseline dataset.' What role does the baseline dataset play?
- A production model's KS statistic drops from 0.65 to 0.41 over 6 months. PSI for input features remains below 0.10. What is the most likely explanation?
- A team deploys a new fraud model using canary deployment with 15% of traffic. The champion continues receiving 85%. After 2 weeks, the canary achieves statistically significantly better precision. What distinguishes canary deployment from A/B testing in this context?
- PROC ASTORE is used to score new observations in SAS. Which PROC ASTORE statement option specifies where the model's input variables and types are defined?
- A blue-green deployment is set up for a SAS Micro Analytic Service scoring endpoint. The 'green' environment contains the new model version. What is the final atomic step that completes the deployment?
- In shadow deployment for a new ML model, what distinguishes it from both A/B testing and canary deployment?
- SAS Model Manager tracks model versions. A data scientist wants to programmatically list all versions of a registered model via the REST API. Which endpoint pattern is correct?
- Model risk management standards (SR 11-7) require that models used in risk decisions be validated by an independent party. In SAS Model Manager, which workflow element supports the independent validation requirement?
- SAS Model Manager's monitoring framework supports model performance monitoring through scheduled monitoring jobs. What type of data is required as input to a performance monitoring job that computes outcome-based metrics (KS, Gini, AUC)?
- A model governance team uses SAS Model Manager to enforce a policy that all models must be validated before promotion to production. Which SAS Model Manager feature enforces this workflow policy?
- A data scientist registers a Python XGBoost model in SAS Model Manager by uploading a ZIP file. After registration, she wants to run the model's score code on a test dataset inside SAS Model Manager. What must be true for the scoring to succeed?
- PROC ASSESS is run with the FITPLOT option on a logistic regression model. What does the FITPLOT calibration plot display?
- A deployed scoring model accumulates 30 days of production predictions. PSI = 0.22 is detected for the predicted score distribution. According to standard PSI thresholds, what action should be taken?
- A model risk management team adds a fairness assessment requirement: models must demonstrate that precision does not differ by more than 5 percentage points across demographic groups. Which SAS Model Manager or SAS Viya workflow step supports this group-stratified fairness metric computation?
- PROC ASTORE DESCRIBE is run on a deployed model. What information is returned in the PROC ASTORE DESCRIBE output?
- In SAS Viya, the SAS Event Stream Processing (ESP) engine is used to score a fraud detection model on streaming transaction data. Which model deployment format does ESP natively support for real-time scoring?
- The PMML format is used to export a logistic regression model from SAS for deployment in a non-SAS scoring environment. Which SAS procedure can export a fitted logistic regression model to PMML?
Domain 6: Deep Learning & NLP with SAS · 52 questions
- A data scientist uses SAS DLPy to build an image classification model. Which deep learning architecture is most appropriate for classifying images of product defects?
- In PROC TEXTMINE, which step in the text mining pipeline transforms raw text documents into a term-by-document matrix?
- A data scientist uses SAS DLPy to define a recurrent neural network for time series forecasting. Which SAS DLPy layer type should be added to capture sequential dependencies?
- In SAS PROC TEXTMINE, what is the purpose of the SVDINIT statement?
- A data scientist trains a text sentiment classifier using SAS and wants to use pre-trained word embeddings. Which SAS capability best supports loading and using pre-trained word vectors such as Word2Vec or GloVe?
- In SAS DLPy, which function is used to instantiate a pre-built convolutional neural network architecture like ResNet-50?
- A data scientist uses SAS DLPy to train a CNN on an image dataset stored in CAS. After training, they want to apply the model to score new images. Which method is used in SAS DLPy to score new data?
- When using PROC TEXTMINE for topic modeling with LDA (Latent Dirichlet Allocation) in SAS, which statement controls the number of topics to discover?
- A data scientist uses SAS DLPy to build a text classification model using a recurrent neural network. Which sequence of steps correctly sets up the model?
- In PROC TEXTMINE, after running the PARSE step, the resulting term-document frequency matrix is stored in output tables. Which output table contains the frequency of each term in each document?
- SAS DLPy provides the TextClassification class for NLP tasks. When fine-tuning a pre-trained BERT model for sentiment analysis using SAS DLPy, which pre-processing step is required before feeding text to BERT?
- A data scientist uses PROC TEXTMINE to perform sentiment analysis on customer reviews. Which statement or option enables the built-in SAS sentiment analysis capability?
- In SAS DLPy, a data scientist trains a CNN model and wants to visualize which regions of an input image most influenced the model's prediction. Which technique is commonly applied for this purpose?
- A data scientist applies PROC TEXTMINE to a corpus of 50,000 customer emails. After parsing, the term-document matrix has 200,000 unique terms. Most terms appear in only 1-2 documents. Which PARSE statement option most effectively removes these rare terms?
- A data scientist implements a convolutional neural network in SAS DLPy for image classification. After adding Conv2D and pooling layers, they add a Flatten layer. What is the purpose of the Flatten layer?
- In SAS DLPy, a data scientist trains an image classifier and wants to augment the training data to improve generalization. Which DLPy class is used to specify image augmentation transformations?
- In PROC TEXTMINE, after SVD is applied using SVDINIT, what are the resulting SVD dimensions used for in subsequent analysis steps?
- A data scientist uses SAS PROC TEXTMINE to analyze customer support tickets and identify common themes. Which output from PROC TEXTMINE best represents the dominant themes across documents?
- A SAS NLP pipeline applies PROC TEXTMINE to e-commerce product reviews to predict star ratings (1-5). After text parsing and SVD, the data scientist wants to train a gradient boosting classifier on the SVD features. What is the correct workflow?
- In SAS DLPy, a data scientist trains a deep learning model and uses the model.getfeaturemaps() method. What does this method return?
- In SAS Viya, a data scientist uses the textParsing CAS action for natural language processing. Which of the following capabilities does the textParsing action provide that PROC TEXTMINE's PARSE statement does not?
- In SAS Viya, a data scientist wants to perform distributed text analytics on 2 million documents. Which architecture advantage does running PROC TEXTMINE in SAS Viya CAS provide over traditional SAS 9.4?
- A data scientist uses SAS DLPy to apply transfer learning by loading a pre-trained VGG16 model and fine-tuning only the final fully connected layers. Which method call freezes the convolutional base layers during fine-tuning?
- In SAS DLPy, which Python package provides the CAS connection object used when initializing deep learning models?
- A data scientist uses PROC TEXTMINE with the NOPARSE option. What effect does this have on text processing?
- Which DLPy parameter enables early stopping during model.fit() training?
- In SAS Viya Visual Text Analytics, what is the primary role of concept extraction?
- Which SAS DLPy architecture is most appropriate for detecting and localizing multiple objects within a single image?
- Which NLP technique maps each word to a dense vector where semantically similar words are geometrically close in the vector space?
- After training a SAS DLPy model, which method exports it as a SAS ASTORE item for deployment via SAS Model Manager?
- In a PROC TEXTMINE followed by PROC DMDB and PROC NEURAL pipeline, what role does PROC DMDB play?
- In SVD decomposition of a term-document matrix (A = UΣVᵀ), what do the singular values in Σ represent?
- In SAS DLPy, the Functional API is used to build a model with two separate input branches. One branch processes images via Conv2d layers, the other processes tabular data via Dense layers. These two branches are merged before a final output layer. Which DLPy class merges (concatenates) the two branch outputs?
- A SAS DLPy image classification model achieves high training accuracy but poor validation accuracy. A practitioner adds Dropout layers after the dense layers. What role do Dropout layers play during training?
- PROC TEXTMINE is run with the MINCOUNT=5 option in the PARSE statement. What effect does MINCOUNT=5 have on the resulting term-document matrix?
- A SAS Visual Text Analytics pipeline uses topic modeling to discover latent themes in customer feedback. Which algorithm is used by SAS Viya's Text Topic action for probabilistic topic discovery?
- Phonetic matching algorithms like Soundex are used in SAS text processing to group spelling variants. Which scenario best illustrates when Soundex would be more appropriate than edit distance (Levenshtein)?
- In SAS DLPy transfer learning, a ResNet-50 pre-trained on ImageNet is adapted for a 10-class medical image classification task. Which layers should initially be frozen during fine-tuning?
- A SAS NLP pipeline produces a named entity recognition (NER) output table where each row is an entity mention. A data engineer needs to aggregate this to the document level to count how many PERSON entities appear per document. Which SAS step accomplishes this most directly?
- A document processing team uses DLPy object detection to identify and localize table regions in scanned financial report images, requiring bounding box coordinates. Which architecture is most appropriate?
- A SAS text analyst needs to score new documents into an existing latent semantic space created by PROC TEXTMINE's SVD. Which SAS procedure is used for this scoring step?
- A SAS DLPy model is trained using the Functional API with two input branches: one Conv2d branch for image features and one Dense branch for tabular metadata. How are the two separate input layers declared in DLPy's Functional API?
- A DLPy sequential model is saved using model.savetoastore(path='models/', name='frauddlpyv1'). What is the purpose of saving to ASTORE format?
- SAS DLPy's EarlyStopping object monitors validation loss with patience=5. A training run progresses through 25 epochs. What does patience=5 specifically mean?
- PROC TEXTMINE is run with WEIGHT=IDFLOG in the PARSE statement. What weighting scheme does IDFLOG apply to terms in the term-document matrix?
- In SAS Viya Text Analytics, sentiment analysis produces sentence-level sentiment scores. A product review contains 5 sentences: 3 positive, 1 neutral, 1 negative. What does the document-level sentiment score represent?
- A DLPy LSTM model is built for time series anomaly detection. The Recurrent() layer is created with rnntype='LSTM' and outputtype='ENCODING'. What does outputtype='ENCODING' return?
- A SAS analyst uses PROC TEXTMINE to analyze 50,000 product reviews. After parsing, the term-document matrix has 25,000 unique terms. The analyst wants to reduce dimensionality using SVD. Which PROC TEXTMINE option controls the number of SVD dimensions (latent topics) retained?
- A data scientist uses DLPy to fine-tune ResNet-50 for medical image classification. The model.fit() call uses the ImageTable class. What SAS infrastructure must be available for DLPy image model training?
- In SAS Viya's deep learning pipeline, the DLPy ImageTable.loadimages() method loads images from a local directory into a CAS in-memory table. What additional CAS action is typically needed to convert raw loaded images to model input format?
- PROC TEXTMINE processes a collection of legal contracts. The RULES statement is included in the PROC TEXTMINE call. What does the RULES statement enable?
- A SAS DLPy model uses Batch Normalization (BN) layers between Conv2d layers. How do Batch Normalization layers behave differently during inference (scoring) compared to during training?
These questions are original practice material and are NOT actual exam questions or brain-dump content. All vendor marks are trademarks of their respective owners. This site is not affiliated with, endorsed by, or sponsored by the exam vendor.