1|Page
ACMS 30600 TEST EXAM 2 (FULL TEST) || 2025
ACTUAL EXAM ALL QUESTIONS AND 100%
CORRECT ANSWERS GRADE A+|| LATEST AND
COMPLETE UPDATED VERSION 2025||
GUARANTEED PASS!!!
R-squared is a good measure of power for a model only if the sample size is
considerably larger than the number of predictors in the model.
A. True
B. False ANSWER- A. True
The slope estimate for an indicator variable for "color_pink" came out to be 22.
What does this value mean?
A. The value of y will be 22 when pink is included in the model.
B. The value of y will be 22 more than the baseline when color is included in
the model.
C. The value of y will be 22 more than the baseline when someone chooses
pink for the variable color.
D. 22 people like the color pink. ANSWER- C. The value of y will be 22 more
than the baseline when someone chooses pink for the variable color.
Which of the following can be used to check the Normality of the error terms?
A. Make a histogram of the residual values and look for a bell shaped graph.
B. Construct a Normal probability plot and check for a straight line pattern.
C. Conduct a formal test of hypothesis using the Shapiro-Wilk test. (The null
hypothesis is Normality.)
,2|Page
D. All of the above ANSWER- D. All of the above (histogram, Normal
probably plot (qqnorm/qqline), hypothesis test with Shapiro-Wilk)
Which of the following is a solution when non-constant variance is present in the
diagnostic plot?
A. Take the natural logarithm of the response variable.
B. Take the square root of the predictor variable.
C. Unequal variances are not a big problem and should be ignored.
D. Both a and b but not c ANSWER- A. Take the natural logarithm of the
response variable.
Which of the following is true about a model utility test?
A. This is a test to see if any of the model parameters are non-zero.
B. The numerator degrees of freedom are k and the denominator degrees of
freedom are n - (k + 1).
C. This is a test to see if a subset of the model parameters are not needed.
D. Both a and b but not c ANSWER- D. Both a and b but not c (if any model
parameters are non-zero, numerator df = k, denominator df = n - (k + 1))
Which of the following is true about model selection?
A. We want models with low AIC and high adjusted R-squared.
B. We want models with low AIC and low adjusted R-squared.
C. We want models with high AIC and high adjusted R-squared.
,3|Page
D. We want models with high AIC and low adjusted R-squared. ANSWER- A.
We want models with low AIC and high adjusted R-squared.
Which of the following is true about principal components analysis (PCA)?
A. It is a technique used for reducing the dimensions of the data.
B. It is a technique used to find rescaled linear combinations of the predictor
variables to reduce collinearity.
C. It provides parameter estimates that are more interpretable than multiple
linear regression.
D. Both a and b but not c ANSWER- D. Both a and b but not c (reduces
dimensions of the data, finds rescaled linear combinations of predictor variables to
reduce collinearity)
Which of the following is true when testing nested models?
A. The denominator degrees of freedom should be k - g, the number of
predictors being tested for removal.
B. The hypotheses are the same as the model utility test.
C. If we fail to reject the null hypothesis, we should use the nested model.
D. Both a and c but not b ANSWER- C. If we fail to reject the null hypothesis,
we should use the nested model.
Why is it advised to not test whether individual model parameters are significant?
A. These are done with t-statistics, which are not as reliable as F-statistics.
, 4|Page
B. The slope estimates for each parameter are only accurate when all other
predictors in the model are held constant.
C. If more than one parameter is tested in sequence, you increase the risk of
committing a type 1 error.
D. Both b and c but not a ANSWER- D. Both b and c but not a (only accurate
when all other predictors are held constant, more than one tested in sequence ->
higher risk of type 1 error)
What are the effects of multicollinearity in our model?
A. It will artificially inflate the standard errors of the parameter estimates.
B. It will decrease the value of our test statistics.
C. It will increase the value of our p-values.
D. All of the above ANSWER- D. All of the above (artificially inflates
standard errors, decreases value of test statistics, increases value of p-values)
What is MSE?
A. the mean standard error
B. the estimated random error variance
C. sum(y_i - yhat_i)^2 / (n - (k + 1))
D. All of the above ANSWER- D. All of the above (mean standard error,
estimated random error variance, sum(y_i - yhat_i)^2 / (n - (k + 1)))
When conducting the Durbin-Watson test for independence, what does a test
statistic near 0 mean?
A. The residuals are positively correlated.
B. The residuals are negatively correlated.
ACMS 30600 TEST EXAM 2 (FULL TEST) || 2025
ACTUAL EXAM ALL QUESTIONS AND 100%
CORRECT ANSWERS GRADE A+|| LATEST AND
COMPLETE UPDATED VERSION 2025||
GUARANTEED PASS!!!
R-squared is a good measure of power for a model only if the sample size is
considerably larger than the number of predictors in the model.
A. True
B. False ANSWER- A. True
The slope estimate for an indicator variable for "color_pink" came out to be 22.
What does this value mean?
A. The value of y will be 22 when pink is included in the model.
B. The value of y will be 22 more than the baseline when color is included in
the model.
C. The value of y will be 22 more than the baseline when someone chooses
pink for the variable color.
D. 22 people like the color pink. ANSWER- C. The value of y will be 22 more
than the baseline when someone chooses pink for the variable color.
Which of the following can be used to check the Normality of the error terms?
A. Make a histogram of the residual values and look for a bell shaped graph.
B. Construct a Normal probability plot and check for a straight line pattern.
C. Conduct a formal test of hypothesis using the Shapiro-Wilk test. (The null
hypothesis is Normality.)
,2|Page
D. All of the above ANSWER- D. All of the above (histogram, Normal
probably plot (qqnorm/qqline), hypothesis test with Shapiro-Wilk)
Which of the following is a solution when non-constant variance is present in the
diagnostic plot?
A. Take the natural logarithm of the response variable.
B. Take the square root of the predictor variable.
C. Unequal variances are not a big problem and should be ignored.
D. Both a and b but not c ANSWER- A. Take the natural logarithm of the
response variable.
Which of the following is true about a model utility test?
A. This is a test to see if any of the model parameters are non-zero.
B. The numerator degrees of freedom are k and the denominator degrees of
freedom are n - (k + 1).
C. This is a test to see if a subset of the model parameters are not needed.
D. Both a and b but not c ANSWER- D. Both a and b but not c (if any model
parameters are non-zero, numerator df = k, denominator df = n - (k + 1))
Which of the following is true about model selection?
A. We want models with low AIC and high adjusted R-squared.
B. We want models with low AIC and low adjusted R-squared.
C. We want models with high AIC and high adjusted R-squared.
,3|Page
D. We want models with high AIC and low adjusted R-squared. ANSWER- A.
We want models with low AIC and high adjusted R-squared.
Which of the following is true about principal components analysis (PCA)?
A. It is a technique used for reducing the dimensions of the data.
B. It is a technique used to find rescaled linear combinations of the predictor
variables to reduce collinearity.
C. It provides parameter estimates that are more interpretable than multiple
linear regression.
D. Both a and b but not c ANSWER- D. Both a and b but not c (reduces
dimensions of the data, finds rescaled linear combinations of predictor variables to
reduce collinearity)
Which of the following is true when testing nested models?
A. The denominator degrees of freedom should be k - g, the number of
predictors being tested for removal.
B. The hypotheses are the same as the model utility test.
C. If we fail to reject the null hypothesis, we should use the nested model.
D. Both a and c but not b ANSWER- C. If we fail to reject the null hypothesis,
we should use the nested model.
Why is it advised to not test whether individual model parameters are significant?
A. These are done with t-statistics, which are not as reliable as F-statistics.
, 4|Page
B. The slope estimates for each parameter are only accurate when all other
predictors in the model are held constant.
C. If more than one parameter is tested in sequence, you increase the risk of
committing a type 1 error.
D. Both b and c but not a ANSWER- D. Both b and c but not a (only accurate
when all other predictors are held constant, more than one tested in sequence ->
higher risk of type 1 error)
What are the effects of multicollinearity in our model?
A. It will artificially inflate the standard errors of the parameter estimates.
B. It will decrease the value of our test statistics.
C. It will increase the value of our p-values.
D. All of the above ANSWER- D. All of the above (artificially inflates
standard errors, decreases value of test statistics, increases value of p-values)
What is MSE?
A. the mean standard error
B. the estimated random error variance
C. sum(y_i - yhat_i)^2 / (n - (k + 1))
D. All of the above ANSWER- D. All of the above (mean standard error,
estimated random error variance, sum(y_i - yhat_i)^2 / (n - (k + 1)))
When conducting the Durbin-Watson test for independence, what does a test
statistic near 0 mean?
A. The residuals are positively correlated.
B. The residuals are negatively correlated.
