1. Data science sits at the intersection of____

a. computer science and data skills

b. math and statistics knowledge, inference and general accounting knowledge

c. computer science and general accounting knowledge

d. computer science and data skills, math and statistics knowledge, and substantive expertise in a particular area of interest or domain such as industry and management accounting knowledge

2. Which of the following is a true statement about the data science framework?

a. The data science framework is the same as the five-step decision-making process

b. The management accountants defer decisions and actions throughout the framework to the data scientists

c. The data science framework is a seven-step decision-making process for applying machine learning techniques in business situations.

d. None of these statements are true regarding the data science framework.

3. Any activity that reveals deeper insight into a dataset is known as ________.

a. Exploratory data analysis

b. Evaluating the data

c. Target feature selection

d. Predictive model selection

4. Target leakage refers to data _______.

a. that are included in the dataset but are duplicates

b. that are private but are inadvertently included in the dataset

c. that are not available at the time of the analysis that should be included.

d. that are not available at the time of the analysis that should be excluded.

5. A decision tree is a technique for ______.

a. modeling focusing on the functional form of the relationship between feature variables and the target variable.

b. segmenting the target variable into different regions based on a set of rules

c. segmenting the feature variables into different regions based on a set of rules

d. segmenting the feature variables into different regions based on the information obtained from the management accountant.

6. Which of the following is false regarding the refining of the decision tree?

a. Overfitting occurs when a model does not adhere closely to the specific details of a dataset such that it captures noise from random change, making it less effective at accurately classifying observations from a new dataset.

b. Pruning is a technique in which the tree is not grown to its full size, but instead is only allowed to grow to a certain depth.

c. Overfitting occurs when a model adheres too closely to the specific details of a dataset such that, in addition to signal, it captures noise from random chance, making it less effective at accurately classifying observations from a new dataset.

d. Overfitting is an important concept for management accountants to recognize when using data analytic techniques.

7. Your full decision tree shows a likelihood value of .07 while the pruned tree shows a likelihood value of .11. Which tree is likely to predict most accurately?

a. the full tree

b. the pruned tree

c. neither

8. Feedback loops and performance on cross-validation data are two important principles that guide ________.

a. the decision of which management accountant to hire

b. the determination of which dataset to use

c. the decision about which should be the target variable

d. the decision about which model to use

9. Which ROC curve suggests a higher level of accuracy?

a. A curve that moves diagonally from bottom left to upper right

b. A curve that moves horizontally near the bottom of the chart

c. A curve that hugs the y-axis and then moves horizontally across the top of the chart

d. A curve that hugs the x-axis and then moves vertically on the right of the chart

10. When we use data science models, as opposed to models that assume a functional relationship, qualitative assessments may safely be ignored.

a. True

b. False