2

9/22/2017 SAS Certified Statistical Business Analyst - SAS 9: Regression and Modeling Credential | Analytics Exam
SAS Certified Statistical Business Analyst - SAS 9: Regression and

Modeling Credential
You got 0 of 60 possible points.

Your score: 0 %
Question Results
Question: Score 0 of 1
(skipped)
In order to perform honest assessment on a predictive model, what is an acceptable
division between training, validation, and testing data?
Response:
Training: 50% Validation: 0% Testing: 50%
(skipped)
The data set WORK.POWERBALL contains a binary target variable name WINNER. You
want to split the data into a training and a validation data set, with 2/3 of the data set aside for training.
The code is started below.
proc surveyselect noprint

data = work.powerball
samprate=.6667
out=work.sample
seed=10974635;
strata winner;
run;
What must be added to this program for it to perform the described task?
Response:
An OUTPUT statement and a SPLITSAMP statement in PROC SURVEYSELECT.
An SPLITSAMP statement in PROC SURVEYSELECT.
http://www.analyticsexam.com/node/3247/take 1/36
An OUTALL option in PROC SURVEYSELECT and a DATA step.
A PROPORTION option in PROC SURVEYSELECT.
(skipped)
This question will ask you to provide missing code segments.
A logistic regression model was fit on a data set where 40% of the outcomes were events (TARGET=1)
and 60% were non-events (TARGET=0). The analyst knows that the population where the model will be
deployed has 5% events and 95% non-events. The analyst also knows that the company's profit margin
for correctly targeted events is nine times higher than the company's loss for incorrectly targeted non-
event.
Given the following SAS program:
What X and Y values should be added to the program to correctly score the data?
Response:
X=40, Y=10
X=.05, Y=10
X=.05, Y=.40
X=.10.Y=05
(skipped)
An analyst, using the GLM procedure, determines that there is a significant interaction
between two categorical predictors:
- income (Low, Medium, High)

- gender (M, F)
The analyst is interested in testing the effect of gender within each level of income. Which GLM
procedure statement will generate these tests?
Response:
lsmeans Gender*Income / slice=Gender;
lsmeans Income / slice=Gender;
lsmeans Gender*Income / slice=Income;
lsmeans Gender / slice=Income;
(skipped)
A financial services manager wants to assess the probability that certain clients will
default on their Home Equity Line of Credit (HELOC). A former employee left the code listed below.
The training data set is named HELOC, while a similar data set of more recent clients is
named RECENT_HELOC. Which SAS data steps will calculate the predicted probability of default on
recent clients? (Choose two.)
Response:
Option A
Option B
Option C
Option D
(skipped)
Refer to the confusion matrix:
Calculate the sensitivity. (0 - negative outcome, 1 - positive outcome)

You can use a calculator if needed.
Response:
25/48
58/102
25/B9
58/81
(skipped)
The question will ask you to provide a missing statement. Given the following SAS
program:
proc logistic data = MYDIR.DEFAULT_DATA des;

model Purchase = Money Acct _type Debt Employment;
<insert statement here>
run;
Which SAS statement will complete the program to correctly score the data set NEW_DATA?
Response:
Scoredata=MYDIR,NEW DATA out=scores;
Scoredata data=MYDIR.NEW_DATA out=scores;
Scoredata data=MYDIR.NEW_DATA output=scores;
Scoredata=HYDIR.NEU_DATA output=scores;
(skipped)
Refer to the REG procedure output:
An analyst has selected this model as a champion because it shows better model fit than a competing
model with more predictors.
Which statistic justifies this rationale?
Response:
R-Square
Coeff Var
Adj R-Sq
Error DF
(skipped)
A non-contributing predictor variable (Pr > |t| =0.658) is added to an existing multiple
linear regression model. What will be the result?
Response:
An increase in R-Square
A decrease in R-Square
A decrease in Mean Square Error
No change in R-Square
(skipped)
Refer to the following exhibit:
What is a correct interpretation of this graph?
Response:
The association between the continuous predictor and the binary response is quadratic.
The association between the continuous predictor and the log-odds is quadratic.
The association between the continuous predictor and the continuous response is quadratic.
The association between the binary predictor and the log-odds is quadratic.
(skipped)
A financial analyst wants to know whether assets in portfolio A are more risky (have
higher variance) than those in portfolio B. The analyst computes the annual returns (or percent changes)
for assets within each of the two groups and obtains the following output from the GLM procedure:
Which conclusion is supported by the output?
Response:
Assets in portfolio A are significantly more risky than assets in portfolio B.
Assets in portfolio B are significantly more risky than assets in portfolio A.
The portfolios differ significantly with respect to risk.
The portfolios do not differ significantly with respect to risk.
(skipped)
Refer to the lift chart:
At a depth of 0.1, Lift = 3.14. What does this mean?
Response:
Selecting the top 10% of the population scored by the model should result in 3.14 times more
events than a random draw of 10%.
Selecting the observations with a response probability of at least 10% should result in 3.14
times more events than a random draw of 10%.
Selecting the top 10% of the population scored by the model should result in 3.14 timesgreater accuracy
than a random draw of 10%.
Selecting the observations with a response probability of atleast 10% should result in 3.14times greater
accuracy than a random draw of 10%.
(skipped)
When selecting variables or effects using SELECTION=BACKWARD in the LOGISTIC
procedure, the business analyst's model selection terminated at Step 3.
What happened between Step 1 and Step 2?
Response:
DF increased.
AIC increased.
Pr > Chisq increased.
- 2 Log L increased.
(skipped)
Refer to the exhibit.
Based on the control plot, which conclusion is justified regarding the means of the response?
Response:
All groups are significantly different from each other.
2XL is significantly different from all other groups.
Only XL and 2XL are not significantly different from each other.
No groups are significantly different from each other.
(skipped)
Logistic regression candidate input variables have missing values. In a predictive
modeling project, what is the primary reason for imputing missing values before fitting a logistic
regression model to the data using PROC LOGISTIC?
Response:
PROC LOGISTIC performs complete case analysis, and even a few missing values per variable can
cause an enormous loss of data.
Imputing missing values using PROC MI will reduce the bias of statistical tests for model parameters.
Imputing missing values using PROC STDIZE with mean imputation will decrease the variance of inputs
having missing values.
Replacing the missing values with some reasonable value will lower the type I error for the statistical
tests of significance of the predictor variables.
(skipped)
A confusion matrix is created for data that were oversampled due to a rare target. What
values are not affected by this oversampling?
Response:
Sensitivity and PV+
Specificity and PV-
PV+ and Pv-
Sensitivity and Specificity
(skipped)
An analyst investigates Region (A, B, or C) as an input variable in a logistic regression
model. The analyst discovers that the probability of purchasing a certain item when Region = A is 1.
What problem does this illustrate?
Response:
Collinearity
Influential observations
Quasi-complete separation
Problems that arise due to missing values
(skipped)
Which SAS program will divide the original data set into 60% training and 40% validation
data sets, stratified by county?
Response:
Option A
Option C
Option B
Option D
(skipped)
How do you get PROC TTEST to display the test for equal variance?
Response:
Use the option EV.
Use the MEANS statement with a HOVTEST option.
Request a plot of the residuals
Use a CLASS statement.
(skipped)
Identify the correct SAS program for fitting a multiple linear regression model with
dependent variable (y) and four predictor variables (x1-x4).
Response:
proc reg data=SASUSER.MLR;

var y x1 x2 x3 x4;
model y = x1-x4;
run;

model y = x1-x4;
run;

model y = x1;
model y = x2;
model y = x3;
model y = x4;
run;

model y = x1 x2 x3 x4 /solution;
run;
(skipped)
Home prices in five regions of a city are compared by an analyst. The following is the
output from the GLM procedure:
Based on this table, what was the analyst able to conclude?
Response:
At least two of the regions have different mean prices.
All of the regions have different mean prices.
At least two of the regions have the same mean price.
All of the regions have the same mean price.
(skipped)
An analyst selects a model as a champion because it shows better model fit than a
competing model with more predictors. Which statistic justifies this rationale?
Response:
R-Square
Model Sums of Squares
AIC
Coefficient of Determination
(skipped)
Which of the following describes a concordant pair of observations in the LOGISTIC
procedure?
Response:
An observation with the event has an equal probability as another observation with the event.
An observation with the event has a lower predicted probability than the observation without the event.
An observation with the event has an equal predicted probability as the observation without the event.
An observation with the event has a higher predicted probability than the observation without the event
(skipped)
Refer to the exhibit:
For the ROC curve shown, what is the meaning of the area under the curve?
Response:
percent concordant plus percent tied
percent concordant plus (.5 * percent tied)
percent concordant plus (.5 * percent discordant)
percent discordant plus percent tied
(skipped)
Refer to the following odds ratio table:
What is a correct interpretation of the estimate?
Response:
The odds of the event are 1.142 greater for each one dollar increase in salary.
The odds of the event are 1.142 greater for each one thousand dollar increase in salary.
The probability of the event is 1.142 greater for each one dollar increase in salary.
The probability of the event is 1.142 greater for each one thousand dollar increase in salary.
(skipped)
The Intercept estimate is interpreted as:
Response:
The predicted value of the response when all the predictors are at their current values.
The predicted value of the response when all predictors are at their means.
The predicted value of the response when all predictors = 0.
The predicted value of the response when all predictors are at their minimum values.
(skipped)
Consider scoring new observations in the SCORE procedure versus the SCORE statement
in the LOGISTIC procedure. Which statement is true?
Response:
The SCORE statement in the LOGISTIC procedure returns only predicted probabilities, whereas the
SCORE procedure returns only predicted logits.
The SCORE statement in the LOGISTIC procedure returns only predicted logits, whereas the SCORE
procedure returns only predicted probabilities.
Unlike the SCORE procedure, the SCORE statement in the LOGISTIC procedure produces both
predicted probabilities and predicted logits.
The SCORE procedure and the SCORE statement in the LOGISTIC procedure produce the same
output.
(skipped)
Identify the correct SAS program for fitting a multiple linear regression model with
dependent
variable (y) and four predictor variables (x1-x4).
Response:
Option A
Option B
Option C
Option D
(skipped)
Given the following LOGISTIC procedure:
proc logistic data = MYDIR.CONVERT des outest=OUTFILE_1;

model Attrite = Calls Plan Billing_code;
score data=MYDIR.NEM ATTRITE DATA out=OUTFILE 2;
run;
What is the difference between the datasets OUTFILEJ and OUTFILE_2?
Response:
OUTFILE_1 contains the final parameter estimates while OUTFILE_2 contains the newly scored
probabilities.
OUTFILE_1 contains the model goodness of fit statistics while OUTFILE_2 contains the newly scored
probabilities
OUTFILE_1 contains the model goodness of fit statistics while OUTFILE_2 contains the newly scored
logits.
OUTFILEJ contains the final parameter estimates and Wald Chi-Square values while OUTFILE_2
contains the newly scored probabilities.
(skipped)
A logging company is studying the germination of pine trees. A programmer developed a
logistic regression model in hopes of predicting the probability of seed germination given data from a
particular location.
She plans on using PROC PLM on the item store saved within PROC LOGISTIC to assist with scoring the
new observations with the developed model.
Complete the PROC PLM code that would produce predicted seed germination probabilities for each
observation.
proc plm restore=pinemodel;

score data=newobs out=scored_pine /
;
title 'Predictions of Pine Germination';
run;
Response:
ULINK
ILINK
YLINK
ZLINK
(skipped)
Which SAS program will detect collinearity in a multiple regression application?
Response:
Option B
Option A
Option C
Option D
(skipped)
Refer to the ROC curve:
As you move along the curve, what changes?
Response:
The priors in the population
The true negative rate in the population
The proportion of events in the training data
The probability cutoff for scoring
(skipped)
What does the reference line at lift = 1 corresponds to?
Response:
The predicted lift for the best 50% of validation data cases
The predicted lift if the entire population is scored as event cases
The predicted lift if none of the population are scored as event cases
The predicted lift if 50% of the population are randomly scored as event cases
(skipped)
Which SAS program will correctly use backward elimination with BIC selection criterion
within the GLMSELECT procedure?
Response:
proc GLMSELECT data=SASUSER.MLR;

model y = x1-x10 /selection=backward select=bic;
run;

model y = x1-x10 /selection=backward choose=bic;
run;

model y = x1-x10 /select=backward selection=bic;
run;

model y = x1-x10 /select=backward choose=bic;
run;
(skipped)
An analyst is screening for irrelevant variables by estimating strength of association
between each input and the target variable. The analyst is using Spearman correlation and Hoeffding's D
statistics in the CORR procedure.
What would likely cause some inputs to have a large Hoeffding and a near zero Spearman statistic?
Response:
nonmonotonic association between the variables
linear association between the variables
monotonic association between the variables
no association between the variables
(skipped)
Screening for non-linearity in binary logistic regression can be achieved by visualizing:
Response:
A scatter plot of binary response versus a predictor variable.
A trend plot of empirical logit versus a predictor variable.
A logistic regression plot of predicted probability values versus a predictor variable.
A box plot of the odds ratio values versus a predictor variable.
(skipped)
An analyst fits a logistic regression model to predict whether or not a client will default on
a loan.
One of the predictors in the model is agent, and each agent serves 15-20 clients each. The model fails to
converge.
The analyst prints the summarized data, showing the number of defaulted loans per agent. See the
partial output below:
What is the most likely reason that the model fails to converge?
Response:
There is quasi-complete separation in the data.
There is collinearity among the predictors.
There are missing values in the data.
There are too many observations in the data.
(skipped)
An analyst has determined that there exists a significant effect due to region. The analyst
needs to make pairwise comparisons of all eight regions and wants to control the experimentwise error
rate.
Which GLM procedure statement would provide the correct output?
Response:
lsmeans Region / pdiff=all adjust=dunnett;
lsmeans Region / pdiff=all adjust=tukey;
lsmeans Region / pdiff=all adjust=lsd;
lsmeans Region / pdiff=all adjust=none;
(skipped)
An analyst examined logistic regression models for predicting whether a customer would make a
purchase. The ROC curve displayed summarizes the models. Using the selected model and the analyst's
decision rule, 25% of the customers who did not make a purchase are incorrectly classified as
purchasers.
Response:
About 25% of the customers who did make a purchase are correctly classified as making a purchase.
About 85% of the customers who did make a purchase are correctly classified as making a
purchase.
(skipped)
An analyst investigates Region (A, B, or C) as an input variable in a logistic regression model. The
analyst discovers that the probability of purchasing a certain item when Region = A is 1.
What problem does this illustrate?
Response:
collinearity
influential observations
quasi-complete separation
problems that arise due to missing values
(skipped)
Which variable shows the highest evidence of non-linear association?
Response:
MortgageAmt
AccountBal
Online
Tenure
(skipped)
Given alpha=0.02, which conclusion is justified regarding percentage of body fat, comparing small (S),
medium (M), and large (L) wrist sizes?
Response:
Medium wrist size is significantly different than small wrist size.
Large wrist size is significantly different than medium wrist size.
Large wrist size is significantly different than small wrist size.
There is no significant difference due to wrist size.
(skipped)
What is a drawback to performing data cleansing (imputation, transformations, etc.) on
raw data prior to artitioning the data for honest assessment as opposed to performing the data
cleansing after partitioning the data?
Response:
It violates assumptions of the model.
It requires extra computational effort and time.
It omits the training (and test) data sets from the benefits of the cleansing methods.
There is no ability to compare the effectiveness of different cleansing methods.
(skipped)
Given the following GLM procedure output:
Which statement is correct at an alpha level of 0.05?
Response:
Gender should not be removed due to its involvement in the significant interaction.
School*Gender should be removed because it is non-significant.
Gender should be removed because it is non-significant.
School should be removed because it is significant.
(skipped)
Including redundant input variables in a regression model can:
Response:
Stabilize parameter estimates and increase the risk of over fitting.
Destabilize parameter estimates and increase the risk of over fitting.
Stabilize parameter estimates and decrease the risk of overfitting.
Destabilize parameter estimates and decrease the risk of overfitting.
(skipped)
The standard form of a linear regression model is:
Which statement best summarizes the assumptions placed on the errors?
Response:
The errors are independent, normally distributed with zero mean and constant variance.
The errors are correlated, normally distributed with constant mean and zero variance.
The errors are correlated, normally distributed with zero mean and constant variance.
The errors are independent, normally distributed with constant mean and zero variance.
(skipped)
Click on the calculator button to display a calculator if needed.
Response:
0.4115
0.6994
0.5884
0.1372
PROC REG was used to model the relationship between the independent variables and the (skipped)
dependent variable, TestScore. The model and partial output are provided below.
TestScore = β0 + β1*Words1 + β2*Letters1+ β3*Phonics1 + ε
Which of the following is a statistical assumption violation that is evidenced by the displayed output?
Response:
Presence of influential outliers.
Non-linearity among the variables.
Lack of normality of the error.
Presence of non-constant error.
(skipped)
A Human Resource manager fits a logistic regression model with the following
characteristics:
- binary target Hired

- continuous predictor Salary
- categorical predictor Education (levels=1,2,3)
The default odds ratio compares each level against the last class level for the variable Education.
Which SAS program gives parameter estimates for Education that are consistent with the default odds
ratios?
Response:
proc logistic data = MYDIR.EMPLOYMENT descending;

class Education (param=ref ref='3');
model Hired = Salary Education;
run;

class Education;
run;

class Education (ref='3');
run;

class Education Salary (param=ref ref='3');
run;
(skipped)
An analyst compared many different models to predict the binary Purchase variable and
selected one particular model.
What rationale supports this decision?
Response:
The model had the highest c statistic on the training data.
The model had the highest c statistic on the validation data.
The model had the lowest c statistic on the training data.
The model had the lowest c statistic on the validation data.
(skipped)
SAS output from the RSOUARE selection method, within the REG procedure, is shown. The top two
models in each subset are given.
Based on the AIC statistic, which model is the champion model?
Response:
Age Weight RunTime RunPulse MaxPulse
Age Weight RunTime RunPulse RestPulse MaxPulse
RestPulse
RunTime
(skipped)
In partitioning data for model assessment, which sampling methods are acceptable?
(There are two correct answer to the question, plz select two answer")
Response:
Simple random sampling without replacement
Simple random sampling with replacement
Stratified random sampling without replacement
Sequential random sampling with replacement

(skipped)
What are the “predicted values” that result from fitting a one-way analysis of variance (ANOVA)
model?
Response:
within-group sample variances
between-group sample variances
within-group sample means
between-group mean differences
(skipped)
Assume a $10 cost for soliciting a non-responder and a $200 profit for soliciting a
responder. The logistic regression model gives a probability score named P_R on a SAS data set called
VALID. The VALID data set contains the responder variable Pinch, a 1/0 variable coded as 1 for
responder. Customers will be solicited when their probability score is more than 0.05. Which SAS
program computes the profit for each customer in the data set VALID?
Response:
Option A
Option B
Option C
Option D
(skipped)
A company has branch offices in eight regions. Customers within each region are
classified as either "High alue" or "Medium Value" and are coded using the variable name VALUE. In the
last year, the total amount of purchases per customer is used as the response variable. Suppose there is
a significant interaction between REGION and VALUE.
What can you conclude?
Response:
More high value customers are found in some regions than others.
The difference between average purchases for medium and high value customers depends on the
region.
Regions with higher average purchases have more high value customers.
Regions with higher average purchases have more medium value customers.
(skipped)
A linear model has the following characteristics:
A dependent variable (y)

Three continuous predictor variables (x1-x3)
One categorical predictor variable (c1with 3 levels)
Which SAS program fits this model?
Response:
Option D
Option A
Option B
Option C
(skipped)
The following SAS code is submitted:
proc reg data=SASHELP.FISH;

model Weight=Length1 Height Width / selection=adjrsq;
run;
How many possible subset models will be assessed by SAS?
Response:
15
(skipped)
What is the general purpose of the ILINK option in PROC LOGISTIC?
Response:
To transform the parameter estimates of the model from the logit scale back to predicted probabilities.
To transform the estimate of the response from the logit scale back to predicted probabilities.
To transform the parameter estimates of the model from the logit scale back to odds ratios.
To transform the estimate of the response from the logit scale back to the odds ratios.
(skipped)
Select the equivalent LOGISTIC procedure model statements. (There are two correct
answer to the question, pl select two answer)
Response:
Mode1 Purchase * Gender Age Region;
Mode1 Purchase * Gender | Age | Region;
Mode1 Purchase * Gender|Age|Region @1;
Mode1 Purchase * Gender|Age|Region @2;
(skipped)
A predictive model uses a data set that has several variables with missing values. What
two problems can arise with this model? (There are two correct answer to the question, plz select two
answer)
Response:
The model will likely be overfit.
There will be a high rate of collinearity among input variables.
Complete case analysis means that fewer observations will be used in the model building process.
New cases with missing values on input variables cannot be scored without extra data processing.
Money Back Guarantee Testimonial FAQs Privacy Policy Terms and Conditions About Us Join Us
Contact Us

2

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

2

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2

Uploaded by

Copyright:

Available Formats

9/22/2017 SAS Certified Statistical Business Analyst - SAS 9: Regression and Modeling Credential | Analytics Exam

SAS Certified Statistical Business Analyst - SAS 9: Regression and

You got 0 of 60 possible points.

Training: 50% Validation: 0% Testing: 50%

Training: 100% Validation: 0% Testing: 0%

Training: 0% Validation: 100% Testing: 0%

Training: 50% Validation: 50% Testing: 0%

proc surveyselect noprint

An OUTPUT statement and a SPLITSAMP statement in PROC SURVEYSELECT.

An SPLITSAMP statement in PROC SURVEYSELECT.

An OUTALL option in PROC SURVEYSELECT and a DATA step.

A PROPORTION option in PROC SURVEYSELECT.

Given the following SAS program:

- income (Low, Medium, High)

lsmeans Gender*Income / slice=Gender;

lsmeans Income / slice=Gender;

lsmeans Gender*Income / slice=Income;

lsmeans Gender / slice=Income;

Calculate the sensitivity. (0 - negative outcome, 1 - positive outcome)

proc logistic data = MYDIR.DEFAULT_DATA des;

Scoredata=MYDIR,NEW DATA out=scores;

Scoredata data=MYDIR.NEW_DATA out=scores;

Scoredata data=MYDIR.NEW_DATA output=scores;

Which statistic justifies this rationale?

A decrease in Mean Square Error

What is a correct interpretation of this graph?

Which conclusion is supported by the output?

Assets in portfolio A are significantly more risky than assets in portfolio B.

Assets in portfolio B are significantly more risky than assets in portfolio A.

The portfolios differ significantly with respect to risk.

The portfolios do not differ significantly with respect to risk.

At a depth of 0.1, Lift = 3.14. What does this mean?

events than a random draw of 10%.

times more events than a random draw of 10%.

What happened between Step 1 and Step 2?

Pr > Chisq increased.

All groups are significantly different from each other.

2XL is significantly different from all other groups.

No groups are significantly different from each other.

Sensitivity and PV+

Specificity and PV-

PV+ and Pv-

Sensitivity and Specificity

What problem does this illustrate?

Problems that arise due to missing values

Use the option EV.

Use the MEANS statement with a HOVTEST option.

Request a plot of the residuals

Use a CLASS statement.

proc reg data=SASUSER.MLR;

proc reg data=SASUSER.MLR;

proc reg data=SASUSER.MLR;

proc reg data=SASUSER.MLR;

Based on this table, what was the analyst able to conclude?

At least two of the regions have different mean prices.

All of the regions have different mean prices.

At least two of the regions have the same mean price.

All of the regions have the same mean price.

TestScore = β0 + β1Words1 + β2Letters1+ β3*Phonics1 + ε