Solutions Manual – Chapter 1
Solutions to Multiple Choice Questions
1. B
2. D
3. C
4. C
5. C
6. C
7. D
8. B
9. B
10. B
Solutions to Discussion Questions
1. Data analytics is defined as the process of evaluating data with the purpose of
drawing conclusions to address business questions. Indeed, effective Data Analytics
provides a way to search through large structured and unstructured data to identify
unknown patterns or relationships.
A university might learn from the analyzing the demographics of its current set of
students in order to attract its future student recruits. Did they come from cities or
high schools that were close by? Were their parents alumni of the university? Did
they score high on certain parts of the ACT? Were those offered a scholarship more
likely to attend, etc.? Was social media effective in attracting students? By analyzing
this type of data, previously unknown patterns will emerge that will make recruiting
students more effective.
2. There are many potential answers. For example, Monsanto may use mathematical
and statistical models to plot out the best times to plant both male and female plans
and where to plant them to maximize yield.
(https://www.cio.com/article/3221621/analytics/6-data-analytics-success-stories-
an-inside-look.html#tk.cio_rs)
3. There are many potential answers. Accountants might use data analytics to learn
more about their allowance for doubtful accounts by learning which customers pay
or do not pay their receivable balances on a timely basis. This will help make a more
accurate balance of net receivables.
4. There are many potential answers. For example, data analytics associated with
financial reporting may help accountants determine if any of their inventory
obsolete? It may also help the company benchmark on the financial statements and
, financial reporting of other similar companies and understand their accounting
practices to help infer their own.
5. The impact cycle suggests an order of 1) Identifying the Questions; 2) Mastering the
Data; 3) Performing the test plan; 4) Addressing and refining results; 5)
Communicating insights and 6) Tracking outcomes. The cycle starts with a question
and then identifying data and test plan that might address that question. The results
of the data analysis are communicated and tracked which may lead to additional,
possibly more refined questions that then restart the cycle.
6. Data analysis is most effective when a question is identified that needs to be
addressed. That will focus the analysis on which data and which test method might
be most effective in addressing or answering the question.
7. Mastering the data requires one to know what data is available and whether it might
be able to help address the business problem. We need to know everything about
the data, including how to access it, its availability, how reliable it is (if there are
errors), and what time periods it covers to make sure it coincides with the timing of
our business problem, etc.
8. Alibaba uses the profiling data approach to identify potential cases of fraud. Alibaba
has worked to capture fraud signals directly from its extensive database of user
behaviors and its network, then analyzes them in real-time using machine learning
to accurately sort the bad users from the good ones.
9. Facebook uses link prediction to predict a relationship between two people when it
suggests people that one likely knows due to similar other friends, high schools,
college or work locations, etc.
10. While sampling is useful, it is still just that, sampling. By looking at all of the
transactions and testing them in a way that will highlight the ones that are the
biggest dollar items, or are most unusual, that will allow auditors to focus on specific
items that might be of material significance.
11. There are several correct answers. One data approach might be regression analysis
where, given a balance of total accounts receivable held by a firm, how long it has
been outstanding, if they have paid debts in the past all will help predict the
appropriate level of allowance for doubtful accounts for bad debts.
12. The Debt-to-Income ratio might suggest to LendingClub that the person asking for
the loan was simply asking for too big of a loan and they would have little ability to
repay it. The lower the credit score, the less likely the loanee would be able to repay
the loan.
13. There are many other potential predictors of whether the LendingClub would pay a
loan. Here are a few possibilities: What other debt do they have? How much is their
disposable income? Do they have a clean criminal record? Have they had a loan with
LendingClub before and did they repay it?
,Solutions to Problems
Note: Some problems and solutions may be altered in Connect for auto grading
purposes.
Problem 1-1
Here are the predictive attributes and whether they would be applicable to predicting
which loans would be delinquent and which loans will ultimately be fully repaid.
Yes/No Predictive Attributes
No desc (Loan description provided by borrower)
Yes dti (Monthly debt payments to monthly income Ratio)
Yes grade (LC assigned loan grade)
home_ownership (values include Rent, Own, Mortgage,
Yes Other)
No next_pymnt_d (Next scheduled payment date)
No term (The number of payments on the loan)
Yes tot_cur_bal (Total current balance of all accounts)
Problem 1-2
Potential attributes from the RejectStats data dictionary that might help predict loan
acceptance or rejection include the following:
Amount Requested
Risk_Score
Debt-to-Income Ratio
Zip Code
State (Possibly)
Employment Length
, Problem 1-3
Percentage of total loans rejected that live in Arkansas = 1.219%
2,915,918 population in Arkansas divided by USA population of 308,745,538 = 0.9444%
The loan rejection percentage is greater than the percent of the USA population that lives in
Arkansas (per 2010 census), but is reasonably close.
Problem 1-4
Loan
Rejection
State %
0.1329270
CA 8
0.0834441
TX 1
NY 0.0797736
0.0768808
FL 9
0.0440198
PA 1
0.0424642
IL 2
0.0377974
OH 4
0.0370800
NJ 8
0.0368352
GA 7
0.0313147
VA 8
0.0271825
MI 5
0.0267239
NC 3
0.0254782
MA 2
0.0234004
MD 8
0.0214281
AZ 1
0.0195455
MO 9
WA 0.0187585
0.0181232
CO 5
Solutions to Multiple Choice Questions
1. B
2. D
3. C
4. C
5. C
6. C
7. D
8. B
9. B
10. B
Solutions to Discussion Questions
1. Data analytics is defined as the process of evaluating data with the purpose of
drawing conclusions to address business questions. Indeed, effective Data Analytics
provides a way to search through large structured and unstructured data to identify
unknown patterns or relationships.
A university might learn from the analyzing the demographics of its current set of
students in order to attract its future student recruits. Did they come from cities or
high schools that were close by? Were their parents alumni of the university? Did
they score high on certain parts of the ACT? Were those offered a scholarship more
likely to attend, etc.? Was social media effective in attracting students? By analyzing
this type of data, previously unknown patterns will emerge that will make recruiting
students more effective.
2. There are many potential answers. For example, Monsanto may use mathematical
and statistical models to plot out the best times to plant both male and female plans
and where to plant them to maximize yield.
(https://www.cio.com/article/3221621/analytics/6-data-analytics-success-stories-
an-inside-look.html#tk.cio_rs)
3. There are many potential answers. Accountants might use data analytics to learn
more about their allowance for doubtful accounts by learning which customers pay
or do not pay their receivable balances on a timely basis. This will help make a more
accurate balance of net receivables.
4. There are many potential answers. For example, data analytics associated with
financial reporting may help accountants determine if any of their inventory
obsolete? It may also help the company benchmark on the financial statements and
, financial reporting of other similar companies and understand their accounting
practices to help infer their own.
5. The impact cycle suggests an order of 1) Identifying the Questions; 2) Mastering the
Data; 3) Performing the test plan; 4) Addressing and refining results; 5)
Communicating insights and 6) Tracking outcomes. The cycle starts with a question
and then identifying data and test plan that might address that question. The results
of the data analysis are communicated and tracked which may lead to additional,
possibly more refined questions that then restart the cycle.
6. Data analysis is most effective when a question is identified that needs to be
addressed. That will focus the analysis on which data and which test method might
be most effective in addressing or answering the question.
7. Mastering the data requires one to know what data is available and whether it might
be able to help address the business problem. We need to know everything about
the data, including how to access it, its availability, how reliable it is (if there are
errors), and what time periods it covers to make sure it coincides with the timing of
our business problem, etc.
8. Alibaba uses the profiling data approach to identify potential cases of fraud. Alibaba
has worked to capture fraud signals directly from its extensive database of user
behaviors and its network, then analyzes them in real-time using machine learning
to accurately sort the bad users from the good ones.
9. Facebook uses link prediction to predict a relationship between two people when it
suggests people that one likely knows due to similar other friends, high schools,
college or work locations, etc.
10. While sampling is useful, it is still just that, sampling. By looking at all of the
transactions and testing them in a way that will highlight the ones that are the
biggest dollar items, or are most unusual, that will allow auditors to focus on specific
items that might be of material significance.
11. There are several correct answers. One data approach might be regression analysis
where, given a balance of total accounts receivable held by a firm, how long it has
been outstanding, if they have paid debts in the past all will help predict the
appropriate level of allowance for doubtful accounts for bad debts.
12. The Debt-to-Income ratio might suggest to LendingClub that the person asking for
the loan was simply asking for too big of a loan and they would have little ability to
repay it. The lower the credit score, the less likely the loanee would be able to repay
the loan.
13. There are many other potential predictors of whether the LendingClub would pay a
loan. Here are a few possibilities: What other debt do they have? How much is their
disposable income? Do they have a clean criminal record? Have they had a loan with
LendingClub before and did they repay it?
,Solutions to Problems
Note: Some problems and solutions may be altered in Connect for auto grading
purposes.
Problem 1-1
Here are the predictive attributes and whether they would be applicable to predicting
which loans would be delinquent and which loans will ultimately be fully repaid.
Yes/No Predictive Attributes
No desc (Loan description provided by borrower)
Yes dti (Monthly debt payments to monthly income Ratio)
Yes grade (LC assigned loan grade)
home_ownership (values include Rent, Own, Mortgage,
Yes Other)
No next_pymnt_d (Next scheduled payment date)
No term (The number of payments on the loan)
Yes tot_cur_bal (Total current balance of all accounts)
Problem 1-2
Potential attributes from the RejectStats data dictionary that might help predict loan
acceptance or rejection include the following:
Amount Requested
Risk_Score
Debt-to-Income Ratio
Zip Code
State (Possibly)
Employment Length
, Problem 1-3
Percentage of total loans rejected that live in Arkansas = 1.219%
2,915,918 population in Arkansas divided by USA population of 308,745,538 = 0.9444%
The loan rejection percentage is greater than the percent of the USA population that lives in
Arkansas (per 2010 census), but is reasonably close.
Problem 1-4
Loan
Rejection
State %
0.1329270
CA 8
0.0834441
TX 1
NY 0.0797736
0.0768808
FL 9
0.0440198
PA 1
0.0424642
IL 2
0.0377974
OH 4
0.0370800
NJ 8
0.0368352
GA 7
0.0313147
VA 8
0.0271825
MI 5
0.0267239
NC 3
0.0254782
MA 2
0.0234004
MD 8
0.0214281
AZ 1
0.0195455
MO 9
WA 0.0187585
0.0181232
CO 5
