Text Material Preview
SAS CERTIFIED ASSOCIATE: MODELING USING SAS VISUAL STATISTICS Exam A00-485 Questions V8.02 SAS Certified Associate: Modeling Using SAS Visual Statistics Topics - Modeling Using SAS Visual Statistics 1 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss 1.How can you export score code generated by SAS Visual Statistics for a trained model? A. By copying and pasting it into a text editor B. By exporting it as a standalone script file C. By printing it directly from the SAS Visual Statistics interface D. By exporting it as a PDF document Answer: AB 2.Refer to the exhibit: Which is the modeling approach that should be used when fitting the Target Gift Amount variable? A. Linear regression model with Interaction effects. B. Generalized linear model with a Poisson distribution and Identity link. C. Generalized linear model with a Normal distribution and Log Link. D. Logistic regression model. Answer: C 3.When evaluating a Misclassification plot, what does it typically show? A. The distribution of predictor variables in the dataset B. The relationship between two predictor variables C. The trade-off between false positives and false negatives at different thresholds D. The p-values associated with the predictor variables Answer: C 4.In SAS Visual Analytics, which of the following tasks involves converting a 2 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss categorical variable into a numerical one? A. Aggregating a measure B. Creating a dummy variable C. Replacing dirty data D. Transforming a variable Answer: B 5.You perform a logistic regression on a multinomial response variable in SAS Visual Statistics that has 3 levels: Small, Medium, Large. "Large" is specified as the event. Which statement is true? A. The other levels are grouped into one non-event. B. An ordinal logistic regression is performed. C. A multinomial logistic regression is performed. D. The other levels are offset to account for exposure. Answer: A 6.When building a cluster analysis in SAS Visual Statistics, which of the following algorithms is commonly used for partitioning data into clusters? A. Principal Component Analysis (PCA) B. Linear Regression C. K-means D. Decision Tree Answer: C 7.How is a multinomial response variable used in SAS Visual Statistics when building a logistic regression model? A. It is used as a predictor variable. B. It is used as an event variable. C. It is used to define the outcome categories. D. It is not used in logistic regression modeling. Answer: C 8.Assuming the event level is the same for both models, which pair of models can be compared? A) 3 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss B) 4 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss C) 5 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss D) 6 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss A. Option A B. Option B C. Option C D. Option D Answer: B 7 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss 9.In cluster analysis, what does the term "centroid" refer to? A. The total number of data points in a cluster B. The center point of a cluster in the feature space C. The number of variables used in the analysis D. The minimum distance between two clusters Answer: B 10.What is meant by "informative missingness" in the context of linear regression? A. Missing data that is unrelated to the outcome variable B. Missing data that is randomly distributed across observations C. Missing data that provides valuable information about the outcome D. Missing data that leads to multicollinearity among predictors Answer: C 11.How can you interpret a Tree Map in the context of a decision tree? A. It visualizes the distribution of data points in leaf nodes. B. It displays the hierarchical structure of the decision tree. C. It shows the decision boundaries between classes. D. It summarizes the performance metrics of the decision tree. Answer: AB 12.Which modeling technique automatically includes missing values? A. Linear regression B. Decision tree C. Logistic regression D. Generalized linear model Answer: B 13.When creating a Linear Regression, Category Variables can be used for Classification Effects and A. Response B. Continuous Effects C. Group-by D. Frequency Answer: C 8 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss 14.Explain what an icicle plot represents in decision tree analysis. A. It displays the hierarchical structure of the decision tree. B. It visualizes the distribution of data points in each leaf node. C. It shows the feature importance for each variable in the tree. D. It summarizes the performance metrics of the decision tree. Answer: AC 15.Refer to the following exhibit: After creating a decision tree in SAS Visual Statistics, what is the value in deriving a Leaf ID Variable? A. Use the leaf ID data item in filters in other types of visualizations. B. Use the leaf ID to rank order by posterior probabilities. C. Use the leaf ID to identify clusters. D. Use the leaf ID to identify the most important variables. Answer: A 16.In the context of linear regression, what does the term "linearity" refer to? A. The use of a straight line to model relationships B. The absence of any relationship between variables C. The requirement for variables to be categorical 9 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss D. The use of polynomial functions for modeling Answer: A 17.What is one advantage of using SAS Studio to score new data with exported score code? A. It requires no additional software installation. B. It provides advanced visualization capabilities. C. It allows for direct integration with Hadoop clusters. D. It automatically optimizes the scoring process. Answer: A 18.What are some properties that can be defined in a linear regression model? A. Learning rate and number of iterations B. Coefficients, p-values, and R-squared C. Maximum depth and minimum leaf size D. Confusion matrix and accuracy Answer: B 19.What is the primary purpose of scoring functionality in SAS Visual Statistics? A. To evaluate model performance on a test dataset B. To generate code for deploying models in production C. To assess the distribution of predictor variables D. To select the best model for a given dataset Answer: AB 20.How can score code be implemented to score new data using SAS Enterprise Miner? A. By pasting the code into a web browser B. By importing the code into SAS Data Integration Studio C. By creating a scoring node in a SAS Enterprise Miner process flow D. By manually executing the code in a command-line interface Answer: C 21.What are the essentials of logistic regression? A. Modeling binary or ordinal outcomes B. Estimating probabilities of categorical outcomes C. Assuming a linear relationship between predictors and response D. Using the logit link function to model the probability of an event 10 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss Answer: ABD 22.When assigning roles in a GLM, what does the "offset" variable typically represent? A. A predictor variable B. The target variable C. A variable with a known constant coefficient D. A variable used to adjust the intercept Answer: CD 23.How is prediction typically performed in logistic regression? A. By calculating the mean of the response variable B. By estimating the probability of an event for each observation C. By fitting a straight line through the data points D. By maximizingthe likelihood of the response variable Answer: B 24.What properties can be assigned when comparing models in SAS Visual Statistics? A. The number of observations in the dataset B. The variables to include in the model comparison C. The learning rate for gradient boosting models D. The color scheme for visualization Answer: B 25.When implementing score code in SAS Enterprise Guide, which step is typically involved in the process? A. Dragging and dropping the score code file into the project B. Running a SQL query to generate predictions C. Opening the code file and manually executing it D. Converting the code to a macro for automation Answer: AC 26.In a generalized linear model, what is the role of the link function in transforming the linear predictor? A. It determines the distribution of the response variable. B. It scales the predictor variables. C. It ensures that the linear predictor is always positive. 11 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss D. It maps the linear predictor to the range of the response variable. Answer: D 27.A champion model has been selected and you want to export the score code. Score code is exported as: A. PMML B. C C. SAS DS2 D. SAS Data Step Answer: D 28.When specifying Cluster Roles for Cluster Analysis in SAS Visual Statistics, which condition is correct? A. Interaction items are allowed. B. A target variable is required. C. At least two variables are required. D. Category items are not allowed. Answer: C 29.What can you interpret from the Assessment panel in a decision tree analysis? A. The tree's structural properties, such as depth and number of nodes B. The tree's training and validation performance metrics C. The decision boundaries between leaf nodes D. The feature importance for each variable in the tree Answer: B 30.In SAS Visual Analytics, what can you use to exclude selections and filter data in a visualization? A. Filter containers B. Exclude option in filter dialogs C. Filtering using the toolbar options D. Interactive filters in the visualization Answer: BD 31.You would like to see the minimum and maximum values for all of your measures so that you can filter variables as needed. Which is the most efficient way to do that? A. Create aggregated measures using the Min and Max aggregations. 12 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss B. Select a histogram object for each measure. C. Select View Measure Details within the Actions menu to the right of the dataset name. D. Create a calculated item subtracting the Min aggregation from the Max aggregation. Answer: C 32.When would you typically use a linear regression model instead of a generalized linear model (GLM)? A. When the dependent variable is binary B. When there is no assumption of linearity in the data C. When dealing with count data D. When the goal is to model categorical outcomes Answer: B 33.How is the KS Statistic typically used when interpreting an ROC chart? A. To measure the area under the ROC curve B. To assess the balance between sensitivity and specificity C. To identify the optimal prediction cut-off threshold D. To evaluate the overall performance of a classification model Answer: B 34.Interactions can be created and added to a linear regression model. Which statement is true about interactions? A. An interaction can be created for a squared category variable. B. Measures and categories can be combined to create an interaction. C. Each interaction is limited to 3 effects or fewer. D. The maximum number of interactions allowed in a model is 3. Answer: B 35.Your company has a dataset that represents global sales. You are a part of a team of analysts that each have responsibility for a certain region of the world. You decide to create a data source filter to suppress every region but yours. What effect will this have on any new explorations that your teammates create? A. It will delete all observations that do not match your region. B. It will have no effect on any observations in the dataset. C. It will suppress all observations that do not match your region. D. It will suppress all observations that do not match their corresponding region. Answer: B 13 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss 36.Which statement is true for negative binomial and Poisson regression models? A. Poisson regression models are used for count data, and negative binomial models are used for binary responses. B. The canonical link function for Poisson regression is the log, while for negative binomial it is the logit. C. Poisson regression is a special case of negative binomial regression. D. Negative binomial models accommodate negative integers while Poisson regression does not. Answer: C 37.In SAS Visual Statistics, which role is typically assigned to variables in cluster analysis? A. Target B. Predictor C. Input D. Output Answer: C 38.Which statement is true regarding the Cluster Analysis in SAS Visual Statistics? A. An observation can belong to more than one cluster. B. Clusters may overlap in the Cluster Matrix plot. C. SAS Visual Statistics creates a measure variable containing the cluster ID when you derive a Cluster ID variable. D. You can derive a Cluster ID variable by right clicking on the Parallel Coordinates plot. Answer: B 39.How can predicted values be derived and described in terms of predicted probabilities in SAS Visual Statistics? A. By calculating the mean of the predictor variables B. By fitting a linear regression model to the data C. By applying the link function to the linear predictor D. By summing the residuals of the model Answer: C 40.Given a scenario, what fit statistic can be used to select a champion model in model comparison? 14 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss A. The p-value of the intercept term B. The R-squared value C. The AUC (Area Under the Curve) D. The learning rate of the model Answer: C 41.Which statement is TRUE regarding a generalized additive model (GAM) in SAS Visual Analytics? A. GAM assumes a strict linear relationship between the predictors and the response function. B. The roughness penalty controls the balance between goodness of fit and the roughness of the spline curve. C. Specification of a spline effect is optional. D. A larger maximum degrees of freedom for the univariate spline term enforces a less complex fit. Answer: B 42.In linear regression, what does the coefficient of determination (R-squared) measure? A. The degree of multicollinearity among predictors B. The goodness of fit of the model to the data C. The impact of outliers on the model D. The p-value associated with the intercept term Answer: B 43.What information can you interpret from a Cluster Matrix? A. The number of data points in each cluster B. The proximity or similarity between clusters C. The cluster centroids D. The order of cluster creation Answer: BC 44.Refer to the exhibit: 15 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss Based on the Validation Data in the Model Comparison output which conclusions are valid? (Choose two.) A. Logistic Regression model would be selected at lift=20 B. Decision Tree model would be selected at lift=5 C. Logistic Regression model would be selected at lift=5 D. Decision Tree model would be selected at lift=20 Answer: CD 45.When reviewing outlier details in linear regression, what factors should be considered when deciding whether to exclude outliers? A. The magnitude of the outlier and its impact on the model B. The number of outliers in the dataset C. The variable roles assigned to outliers D. The p-values associated with the outliers Answer: A 46.Given a scenario where the response variable represents the time until an event occurs, what distributionand link function might be appropriate for modeling? A. Poisson distribution with a log link function B. Normal distribution with an identity link function C. Exponential distribution with a log link function D. Logistic distribution with a logit link function Answer: C 16 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss 47.Cluster Analysis in SAS Visual Statistics is categorized as which type of machine learning? A. Supervised learning B. Semi-supervised learning C. Unsupervised learning D. Reinforcement learning Answer: C 48.Which model does not produce score code? A. Decision Tree using interactive mode B. Regression using interaction effects C. Regression using the group by option D. Decision Tree using the rapid growth option Answer: A 49.Refer to the exhibit from a linear regression model in SAS Visual Statistics. Based on the table above and assuming a significance level of 0.05, what can be concluded about the linear regression model? A. The Intercept is an important predictor of the response. B. RestPulse is a significant predictor of the response. C. For one one-unit increase in RunTime, there is an expected increase in the response of 2.6287. D. For a .03696 unit decrease in RunPulse, there is an expected one-unit increase in the response. Answer: C 50.Given an Influence Plot generated in a Linear Regression within SAS Visual Statistics, which diagnostic can be used to quantify leverage? A. Likelihood Displacement 17 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss B. AIC C. Adjusted R-Squared D. Concordance Answer: A 51.How can you appropriately change the number of clusters in a k-means cluster analysis? A. By adjusting the learning rate B. By modifying the random seed C. By changing the K-value D. By increasing the number of iterations Answer: C 52.Refer to the exhibit from SAS Visual Statistics: What can be concluded from the lift plot shown in the exhibit? A. The 20% of cases with the highest predicted probability of the modeled event level is expected to capture 2.75 times as many events as a random sample of the same size. B. The response rate for the best 20% is predicted to increase 5 times under the fitted model. C. The 20% of cases with the lowest predicted probability of the modeled event level 18 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss is predicted to capture 2.75 times as many events as a random sample of 20%. D. The response rate for the best 20% is predicted to be 2.75 times greater than the sample fitted in the model. Answer: A 53.What is the primary goal of linear models in statistical modeling? A. To predict categorical outcomes B. To capture non-linear relationships in data C. To establish a linear relationship between variables D. To classify data into distinct groups Answer: C 54.What is the primary benefit of exporting score code for model deployment? A. It allows for model performance assessment on a test dataset. B. It enables scoring of new data without the need for SAS Visual Statistics. C. It provides insights into the distribution of predictor variables. D. It automatically generates new predictor variables. Answer: B 55.In decision tree modeling, what is the purpose of setting a minimum split count? A. To prevent any splits in the tree B. To determine the maximum depth of the tree C. To specify the minimum number of data points required to perform a split D. To control the learning rate of the model Answer: C 56.Refer to the fit summary from in the exhibit below. 19 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss What can be concluded from the fit summary? A. Job Type is not a significant predictor in this model. B. Job Type = Office has no important variables associated with it. C. Debt to income ratio is a significant predictor when Job Type = Mgr. D. Debt to income ratio is a significant predictor when Job Type = Sales. Answer: C 57.Refer to the exhibit: A cluster has been created with four variables selected on the roles tab. Select the set of Parallel Coordinates Properties that would display the image in the exhibit. A) 20 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss B) C) D) 21 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss A. Option A B. Option B C. Option C D. Option D Answer: D 58.Refer to the exhibits: 22 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss 23 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss An analyst has created a cluster model based on the settings in Exhibit 1 and chose to derive a cluster ID variable. Why are there 6 cluster IDs in Exhibit 2 based on these settings? A. Cluster Diagram Visible Roles setting is 6. B. There are observations with missing values. C. Variable Standardization has been chosen. D. Parallel Coordinates Visible Roles setting is 6. Answer: B 59.You have a regression model effect that represents the total amount of sales. In addition to that, you would like to create a model effect that represents the average amount of sales. Which option should you use? A. Create an aggregated measure using the Avg aggregation on total amount of sales. B. Create a calculated item that divides total amount of sales by the total amount of items sold. C. Create a calculated item by duplicating the original model effect and changing its default aggregation to Average. D. Create an aggregated measure using the Sum aggregation of total amount of sales 24 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss divided by the Sum aggregation of total amount of items sold. Answer: C 60.What is one potential disadvantage of excessively deep decision trees? A. Increased interpretability B. Overfitting to the training data C. Faster prediction times D. Greater stability in the model Answer: B 61.During cluster creation, if the Variable Standardization option is selected, where in SAS Visual Statistics can you view the standardization formula? A. score code B. cluster summary table C. cluster matrix plot D. parallel coordinates plot Answer: A 62.When setting proper inputs for the k-means algorithm, what parameter represents the number of clusters to create? A. Learning rate B. Random seed C. Cluster centroid initialization D. K-value Answer: D 63.What information is conveyed by a Summary bar in model assessment? A. The standard deviation of the predictor variables B. The summary statistics of the response variable C. The performance metrics of the model, such as AUC and KS Statistic D. The distribution of residuals in the model Answer: C 64.How can you interpret a Parallel Coordinates plot in cluster analysis? A. It shows the distribution of data within a cluster. B. It visualizes the relationship between clusters. C. It displays the hierarchical structure of clusters. D. It provides a summary of cluster statistics. 25 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss Answer: AB 65.Prior to fitting a regression model, data binning of a continuous effect should be used for what purpose? A. Improve the accuracy of your predictions. B. Reduce the impact of minor observation errors. C. Reduce the cardinality of a continuous effect. D. Gain information due to discretization. Answer: B 66.What is the purpose of reviewing Measure Details when preparing data for a model? A. To check data types and formats of the measures B. To understand the level of measurement for each variable C. To identify missing values in the measuresD. To determine the aggregation method for measures Answer: AB 67.Interactive group-by analysis can be performed for each of the following in Visual Statistics except: A. Logistic regression B. Linear regression C. Decision tree D. Generalized linear model Answer: C 68.Refer to the exhibit: 26 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss A retailer wants to use a decision tree to select one segment to mail 35,000 coupons to existing customers for a new line of organic products. Assuming that customers who have purchased organic products in the past are likely to want new organic products in the future, select the single best Leaf ID for the mailing. A. 18 B. 15 C. 17 D. 12 Answer: A 69.In the below linear regression results display, where would you click to access the parameter estimates? 27 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss Answer: 28 / 30 SAS A 00 -4 85 E xa m Q ue st io ns to E ns ur e You r S uc ce ss 70.What does the Variable Importance feature help with when choosing the best fitting group-by model? A. Identifying the most influential predictors B. Assessing the significance of each predictor C. Evaluating the multicollinearity among predictors D. Understanding the distribution of the target variable Answer: A 29 / 30 Get full version of A00-485 Q&As Powered by TCPDF (www.tcpdf.org) 30 / 30 https://www.killtest.com/SAS-Certified-Associate-Modeling-Using-SAS-Visual-Statistics/A00-485.asp https://www.killtest.com/SAS-Certified-Associate-Modeling-Using-SAS-Visual-Statistics/A00-485.asp http://www.tcpdf.org
