In SAS Visual Statistics, where do CAS in-memory tables that power model objects live?
Cross Functional Tasks
Question · Easy
Overfitting: I memorized the practice test
Short
Which measure level should you assign to a variable that has only two distinct outcomes such as Yes or No?
A continuous numeric variable used as the model target should be assigned which measure level?
Tech Exam Prep Podcast
Listen on Spotify
Podcast
Which variable role tells SAS Visual Statistics that a row's contribution should be replicated by an integer count?
Question · Medium
Which role is used to assign a non-integer importance value that affects variance estimation?
What does the Group By role do when added to a model in SAS Visual Statistics?
After group-by modeling produces hundreds of fits, which view ranks each by-group model on a chosen statistic?
Which interactive technique in SAS Visual Statistics highlights selected points across all linked report objects?
A modeler refits a regression after dragging in a new predictor. Which property of SAS Visual Statistics enables this immediate response?
Which statement about the global filter on a SAS Visual Statistics page is correct?
Which Visual Statistics object performs unsupervised segmentation of observations based on similarity of input variables?
Segmentation
Which clustering algorithm underlies the Cluster object in SAS Visual Statistics?
Which input variables are valid for a Cluster object?
Which option helps reduce the influence of variables measured on different scales in k-means clustering?
After fitting a cluster model, which Visual Statistics view describes how each cluster differs from the overall mean on its inputs?
How can you create a categorical variable that represents cluster membership for use in downstream models?
When deciding the number of clusters k, which heuristic is commonly used in Visual Statistics?
Which split criterion is available in the Visual Statistics decision tree for an interval target?
For a nominal (categorical) target, which split criterion does the Visual Statistics decision tree apply?
Which decision tree control limits how deep the tree can grow from the root?
Which decision tree property prevents creating tiny terminal nodes that may overfit?
Pruning a decision tree primarily improves which property?
What does the Visual Statistics decision tree do with missing values in a predictor by default?
Which decision tree output shows the human-readable IF-THEN logic that defines each leaf?
How can the leaf assignment from a decision tree be reused as a predictor in another model?
Which Visual Statistics object would best segment a sales territory to find natural customer groups using only demographic features?
Which output most directly tells you which inputs are driving a decision tree's predictions?
Which property does NOT directly control decision tree complexity?
Which clustering output is most useful for naming each cluster in a marketing context?
When fitting a cluster object with one extreme outlier present, what is a likely consequence?
Which Visual Statistics model is appropriate when the target is interval (continuous) and you assume a linear additive relationship?
Regression
In a Visual Statistics linear regression, what is the interpretation of the coefficient on a continuous predictor?
Which fit statistic for a linear regression in Visual Statistics indicates the proportion of variance in the target explained by the model?
An informative residual versus predicted plot for a Visual Statistics linear regression should look like what?
Which target type setting is correct for a logistic regression predicting churn (Yes/No)?
Why must you set the event level for a binary logistic regression?
Which target measure level supports multinomial logistic regression in Visual Statistics?
Which Visual Statistics output describes how well a logistic regression separates events from non-events across all probability cutoffs?
What does an AUC value of 0.5 indicate for a binary classification model?
What does the KS statistic (Kolmogorov-Smirnov) shown in the ROC chart measure?
A generalized linear model is preferred over ordinary linear regression when which condition holds?
Which distribution and link combination is most appropriate for a count target such as number of claims?
Which distribution and link combination is the canonical choice for a binary target in a GLM?
Which link function produces a model whose linear predictor is the natural log of the mean?
Which link function is sometimes used as an alternative to logit for binary targets, based on the inverse normal CDF?
Which distribution is most appropriate for a positive, right-skewed continuous target such as waiting times?
Loading more…