Ch. 6 - Data Driven Detection

Question 1

detects control failures and
routine misstatements/ weak
for detecting rare fraud

Answer 1

sampling

Question 2

____ = using data, usually
through the mining of those
data, to identify patterns,
annomolies to find possible fraud symptoms

Answer 2

Data driven fraud detection
- proactive
- hypothesis driven
- full population focused

Question 3

test the entire dataset

Answer 3

full population analysis

Question 4

6 steps to the
Data Analysis Process

Answer 4

1. Understand the business
2. Identify possible fraud that
   could exist (revenue? inventory?)
3. Catalog possible fraud symptoms
   (what would that fraud look like in the data?)
4. Use technology to gather data
   about symptoms
5. Analyze results
6. Investigate symptoms

Question 5

Fraud Symptom categories
(part of step 3)

Answer 5

• Accounting errors
• Internal control weaknesses
• Analytical anomalies
• Extravagant lifestyles
• Behavioral
• Tips and complaints

Question 6

the primary advantage of
the data drive approach is

Answer 6

the investigator takes charge
of the fraud investigation process

Question 7

the primary drawback to
the data driven approach is that

Answer 7

it can be more expensive and time
intensive that the traditional approach

Question 8

accurately predicts for many kinds
of financial data that the first digits
of each group of numbers in a set
of random numbers will conform
to the predicted distribution pattern

Answer 8

Benfords law
- invoce amounts/
expense reports/transaction totals

Question 9

outliers identify cases
that do not match the norm
- how far outside the norm is the case

Answer 9

Z-score
* -1-1 = 68% of population
* -2-2 = 95%
* -3-3 = 97.5%

Question 10

the splitting of complex
data sets into groupings

Answer 10

stratification
(fraud hides in categories:
by vendor/by buyer, etc.

Question 11

a technique that produces
a single number that summarizes
each group

Answer 11

time trend analysis
- fraud escalates

Question 12

detects near identical names

Answer 12

fuzzy matching
- uncocer shell companies/
related party vendors

Question 13

(real-time)
modern systems can

Answer 13

• flag anomalies at transaction time
• apply automated fraud indicators
• detect fraud earlier
• not reactive/now proactive