Question 1

The AIC never penalizes the model for the number of model coefficients; therefore we do not favor a minimum AIC value.

Accepted Answer

A) True 
 B)False

Question 2

Parametric predictive analysis can be used to classify qualitative variables into several groups. The processes might include:

Accepted Answer

A)  logistic regression. 
B)  simple regression. 
C)  vertical discriminate analysis. 
D)  stratification and clustering techniques. 
A)  logistic regression. 
B)  simple regression. 
C)  vertical discriminate analysis. 
D)  stratification and clustering techniques.

Question 3

In logistic regression, the difference between a fitted model and the saturated model can be measured with what test?

Accepted Answer

A)  Deviance 
B)  Pearson 
C)  Hosmer-Lemeshow 
D)  All of the choices are correct. 
A)  Deviance 
B)  Pearson 
C)  Hosmer-Lemeshow 
D)  All of the choices are correct.

Question 4

What is JMP and how is it used in statistical processes?

Accepted Answer

JMP is software that automates

Question 5

Which predictive analytic technique would be best suited to an analysis of Facebook users, age 50-59, who purchase items on Amazon?

Accepted Answer

A)  Neural network modeling due to the large data set with many observations. 
B)  Logistic regression, since the dependent variable is age. 
C)  Logistic regression, since the dependent variable is purchase items on Amazon, or does not purchase items on Amazon. 
D)  Either neural network modeling or logistic regression would work, depending on how close you want your p-value to .05. 
A)  Neural network modeling due to the large data set with many observations. 
B)  Logistic regression, since the dependent variable is age. 
C)  Logistic regression, since the dependent variable is purchase items on Amazon, or does not purchase items on Amazon. 
D)  Either neural network modeling or logistic regression would work, depending on how close you want your p-value to .05.

Question 6

What is a major drawback of neural network modeling?

Accepted Answer

A)  The parameters are typically uninterpretable. 
B)  The parameters are typically interpretable. 
C)  The independent variables cannot be defined. 
D)  The level of significance is below .05, making the results invalid. 
A)  The parameters are typically uninterpretable. 
B)  The parameters are typically interpretable. 
C)  The independent variables cannot be defined. 
D)  The level of significance is below .05, making the results invalid.

Question 7

When we compare the squared differences in linear discriminate analysis we are looking for the larger value to predict the outcome of that data point.

Accepted Answer

A) True 
 B)False

Question 8

A neural network is nonlinear.

Accepted Answer

A) True 
 B)False

Question 9

Which of the following tests can be used to assess whether the logistic regression model is a good fit?

Accepted Answer

A)  Homer-Lemeshow test 
B)  Pearson 
C)  ROC curve 
D)  All of the choices are correct. 
A)  Homer-Lemeshow test 
B)  Pearson 
C)  ROC curve 
D)  All of the choices are correct.

Question 10

A logistic curve initially grows exponentially and

Accepted Answer

A)  is theoretical. 
B)  is a summary of the probabilities. 
C)  is a consistent curve regardless of the data. 
D)  reflects the confidence intervals. 
A)  is theoretical. 
B)  is a summary of the probabilities. 
C)  is a consistent curve regardless of the data. 
D)  reflects the confidence intervals.

Question 11

Logistic regression is used when you want to

Accepted Answer

A)  predict a categorical variable from continuous or categorical variables. 
B)  predict a continuous variable from categorical variables. 
C)  predict any categorical variable from other categorical variables. 
D)  predict a continuous variable from a categorical variable. 
A)  predict a categorical variable from continuous or categorical variables. 
B)  predict a continuous variable from categorical variables. 
C)  predict any categorical variable from other categorical variables. 
D)  predict a continuous variable from a categorical variable.

Question 12

Forward selection with simultaneous validation is when one predictor variable is added at a time to a model where the variable added to the model has the smallest p-value when added to the model.

Accepted Answer

A) True 
 B)False

Question 13

In a neural network, the penalty weight controls the trade-off between ________ and ________.

Accepted Answer

A)  overfitting; underfitting 
B)  overfitting; not fitting 
C)  underfitting; fitting 
D)  fitting; not fitting 
A)  overfitting; underfitting 
B)  overfitting; not fitting 
C)  underfitting; fitting 
D)  fitting; not fitting

Question 14

There are many choices of statistical analysis software available. Several were discussed in the text. Compare and contrast Minitab and JMP, as they relate to logistic regression, discriminate analysis, and neural network modeling.

Accepted Answer

will vary here. Since students were aske

Question 15

The probability of a new employee passing a test is .20. What are the odds of the employee passing the test?

Accepted Answer

A)  .20 
B)  .25 
C)  .80 
D)  .75 
A)  .20 
B)  .25 
C)  .80 
D)  .75

Question 16

You want to add new features to data to which you applied a logistic regression model and got a training accuracy of X and a testing accuracy of Y. Which options are correct, if you consider the remaining parameters are the same?

Accepted Answer

A)  Training accuracy increases and testing accuracy increases or remains the same. 
B)  Training accuracy increases or remains the same. 
C)  Testing accuracy decreases and training accuracy remains the same. 
D)  Testing accuracy increases or remains the same and training accuracy decreases. 
A)  Training accuracy increases and testing accuracy increases or remains the same. 
B)  Training accuracy increases or remains the same. 
C)  Testing accuracy decreases and training accuracy remains the same. 
D)  Testing accuracy increases or remains the same and training accuracy decreases.

Question 17

Review the price reduction question in the text. Different software programs were used to evaluate whether a customer would redeem a coupon or not redeem a coupon. Explain the implications of the logistic regression Minitab results (Figure 16.1), the JMP output of logistic regression (Figure 16.2). How does the output vary? What can you determine from the various methods of evaluating the data?

Accepted Answer

Each student will have differing Answer:

Question 18

Below gives the data concerning (1) the dependent variable Default which equals 1 if a customer defaults on their loan and 0 if they do not; (2) the independent variable Price of Home, which is the price of the home (in tens) and (3) the independent variable First Purchase which equals 0 if the customer has owned a home before and 1 if this is their first home. Identify and interpret the odds ratio estimate for Price of Home.

Accepted Answer

A)  Odds ratio: 4.56789; for each addition $10 spent on a home, the odds of a person defaulting increased 4.56% 
B)  Odds ratio: 4.56789; for each addition $100 spent on a home, the odds of a person defaulting increased 4.56% 
C)  Odds ratio: 4.56789; for each addition $10 spent on a home, a default is 4 times more likely 
D)  Odds ratio: 4.56789; for each addition $1000 spent on a home, a default is 4 times less likely for each additional $1,000 spent 
A)  Odds ratio: 4.56789; for each addition $10 spent on a home, the odds of a person defaulting increased 4.56% 
B)  Odds ratio: 4.56789; for each addition $100 spent on a home, the odds of a person defaulting increased 4.56% 
C)  Odds ratio: 4.56789; for each addition $10 spent on a home, a default is 4 times more likely 
D)  Odds ratio: 4.56789; for each addition $1000 spent on a home, a default is 4 times less likely for each additional $1,000 spent

Question 19

In linear discriminate analysis we compare squared differences to see where a particular data point lies in relation to the categorical outcome.

Accepted Answer

A) True 
 B)False

Question 20

Neural networks are common for large data mining projects to analyze a data set involving millions of observations.

Accepted Answer

A) True 
 B)False

The AIC never penalizes the model for the number of model coefficients; therefore we do not favor a minimum AIC value.

Parametric predictive analysis can be used to classify qualitative variables into several groups. The processes might include:

In logistic regression, the difference between a fitted model and the saturated model can be measured with what test?

What is JMP and how is it used in statistical processes?

Which predictive analytic technique would be best suited to an analysis of Facebook users, age 50-59, who purchase items on Amazon?

What is a major drawback of neural network modeling?

When we compare the squared differences in linear discriminate analysis we are looking for the larger value to predict the outcome of that data point.

A neural network is nonlinear.

Which of the following tests can be used to assess whether the logistic regression model is a good fit?

A logistic curve initially grows exponentially and

Logistic regression is used when you want to

Forward selection with simultaneous validation is when one predictor variable is added at a time to a model where the variable added to the model has the smallest p-value when added to the model.

In a neural network, the penalty weight controls the trade-off between and .

There are many choices of statistical analysis software available. Several were discussed in the text. Compare and contrast Minitab and JMP, as they relate to logistic regression, discriminate analysis, and neural network modeling.

The probability of a new employee passing a test is .20. What are the odds of the employee passing the test?

You want to add new features to data to which you applied a logistic regression model and got a training accuracy of X and a testing accuracy of Y. Which options are correct, if you consider the remaining parameters are the same?

In linear discriminate analysis we compare squared differences to see where a particular data point lies in relation to the categorical outcome.

Neural networks are common for large data mining projects to analyze a data set involving millions of observations.

An Introduction to Business Statistics and Analytics

Descriptive Statistics and Analytics: Tabular and Graphical Methods

Descriptive Statistics and Analytics: Numerical Methods

Probability and Probability Models

Predictive Analytics I: Trees, K-Nearest Neighbors, Naive Bayes,

Discrete Random Variables

Continuous Random Variables

Sampling Distributions

Confidence Intervals

Hypothesis Testing

Statistical Inferences Based on Two Samples

Experimental Design and Analysis of Variance

Chi-Square Tests

Simple Linear Regression Analysis

Multiple Regression and Model Building

Time Series Forecasting and Index Numbers

Nonparametric Methods

Decision Theory

Filters

Exam 16: Predictive Analytics Ii: Logistic Regression, Discriminate Analysis,

The AIC never penalizes the model for the number of model coefficients; therefore we do not favor a minimum AIC value.

Parametric predictive analysis can be used to classify qualitative variables into several groups. The processes might include:

In logistic regression, the difference between a fitted model and the saturated model can be measured with what test?

What is JMP and how is it used in statistical processes?

Which predictive analytic technique would be best suited to an analysis of Facebook users, age 50-59, who purchase items on Amazon?

What is a major drawback of neural network modeling?

When we compare the squared differences in linear discriminate analysis we are looking for the larger value to predict the outcome of that data point.

A neural network is nonlinear.

Which of the following tests can be used to assess whether the logistic regression model is a good fit?

A logistic curve initially grows exponentially and

Logistic regression is used when you want to

Forward selection with simultaneous validation is when one predictor variable is added at a time to a model where the variable added to the model has the smallest p-value when added to the model.

In a neural network, the penalty weight controls the trade-off between ________ and ________.

There are many choices of statistical analysis software available. Several were discussed in the text. Compare and contrast Minitab and JMP, as they relate to logistic regression, discriminate analysis, and neural network modeling.

The probability of a new employee passing a test is .20. What are the odds of the employee passing the test?

You want to add new features to data to which you applied a logistic regression model and got a training accuracy of X and a testing accuracy of Y. Which options are correct, if you consider the remaining parameters are the same?

In linear discriminate analysis we compare squared differences to see where a particular data point lies in relation to the categorical outcome.

Neural networks are common for large data mining projects to analyze a data set involving millions of observations.

An Introduction to Business Statistics and Analytics

Descriptive Statistics and Analytics: Tabular and Graphical Methods

Descriptive Statistics and Analytics: Numerical Methods

Probability and Probability Models

Predictive Analytics I: Trees, K-Nearest Neighbors, Naive Bayes,

Discrete Random Variables

Continuous Random Variables

Sampling Distributions

Confidence Intervals

Hypothesis Testing

Statistical Inferences Based on Two Samples

Experimental Design and Analysis of Variance

Chi-Square Tests

Simple Linear Regression Analysis

Multiple Regression and Model Building

Time Series Forecasting and Index Numbers

Nonparametric Methods

Decision Theory

Filters

In a neural network, the penalty weight controls the trade-off between and .