Unemployment Insurance Integrity ConferenceApril 19, 2010

Forensic TechniquesAnd

Automated Oversight

Brett Baker, PhD, CPA, CISA

2

Overview

• Forensic Techniques and Automated Oversight• Data Mining• Techniques• Equipment and Software• Forensic Approach

Brett M. Baker, PhD, CPA, CISA

3

Forensic Techniques and Automated Oversight

• Definition of Forensic Audit– Audit that specifically looks for financial misconduct, abusive or

wasteful activity. – Close coordination with investigators– More than Computer Assisted Audit Techniques (CAATs)

• Forensic audit is growing in the Federal government– GAO’s Forensic Audit and Special Investigations (FSI)– Department of Defense Data Mining

• Federal outlays are $2 trillion annually– OMB estimates improper payments for Federal government at $98B

(4%)• 100% review using automated business rules versus

statistical sampling– There is a place for both

• Automated Oversight– Continuous monitoring– Quick response

Brett M. Baker, PhD, CPA, CISA

FY2008 Improper Payment Estimates

Brett M. Baker, PhD, CPA, CISA

5

What is Data Mining?

• Refers to the use of machine learning and statistical analysis for the purpose of finding patterns in data sets. – If You Know Exactly What You Are Looking for, Use

Structured Query Language (SQL).– If You Know Only Vaguely What You Are Looking for,

Turn to Data Mining.

• Most often used (up until recently) in marketing and customer analysis

Brett M. Baker, PhD, CPA, CISA

6

Different Levels of Knowledge

Information Summary Reports

KnowledgeDescriptive Analytics

WisdomPredictive Analytics

DataFacts, numbers

ACL, IDEA, MS Access

SAS, SPSS, ACL, IDEA

ClementineIntelligent MinerEnterprise Miner

Brett M. Baker, PhD, CPA, CISA

7

• Can perform the tests wanted, instead of being limited to what technical staff can, or will, provide

• Not limited to just predetermined data formats and/or relationships

• Can create relationships, check calculations and perform comparisons

• Can examine all records, not just a sample• Useful for identifying misappropriation of assets

and fraudulent financial reporting• Allows limitless number of analytical relationships

to be assessed– within large databases– comparing large databases

• Identifies anomalies

Data Analysis Software - Fosters Creativity

Brett M. Baker, PhD, CPA, CISA

8

Common Data Analysis Tests and Techniques

• Join• Summarization• Corrupt data (conversion)• Blank fields (noteworthy if field is mandatory)• Invalid dates• Bounds testing• Completeness• Uniqueness• Invalid codes• Unreliable computed fields• Illogical field relationships• Trend analysis• Duplicates

Brett M. Baker, PhD, CPA, CISA

99

• Use Tight Selection Criteria– E.g. Vendor = “Smith Company”

• Use Looser Selection Criteria– E.g. Vendor = “*Smith*”

• Finds Smith Company, Smith Co., The Smith Manufacturing Company, etc.

You Know Exactly What You’re Looking For e.g. All You Know Exactly What You’re Looking For e.g. All Payments to a Particular VendorPayments to a Particular Vendor

Simple Queries

1010

• Use Tight Selection Criteria– E.g. Payments With Same Contract #,

Invoice # and Invoice Amount

– Few False Positives

• Use Looser Selection Criteria– E.g. Payments With Same Contract # and

Invoice # or Same Contract # and Invoice Amount

– More False Positives– More Detected Duplicates

You Know The Specific Condition You’re Looking For e.g. Duplicate Payments

Complex Queries

1111

• Subject Matter Experts Can Describe Indicators– Translate Indicators to Detection Logic

• Apply Indicators Against Population– Many False Positives

• Combine Indicators – Reduce False Positives

You Only Know The General Condition You’re Looking For e.g. Fraud

Sophisticated Solutions

1 5 7

2 8 4

12 15 3

Combine Indicators

Control Charts

Brett M. Baker, PhD, CPA, CISA

Normal Activity AnomalousActivity

AnomalousActivity

Frequency Distribution

Brett M. Baker, PhD, CPA, CISA

DisbursingTransactions

Vendor Table

VendorsNot Paid

Yet

VendorsPaid andIn Vendor

Table

VendorsPaid but not

In VendorTable

Comparing Data Files(Three-Bucket Theory)

Brett M. Baker, PhD, CPA, CISA

15

Hardware and Software Applications

• Hardware– SQL servers– Mainframe (QMF)– Docking stations– Terminal server

• Software Applications– Data mining and predictive analytics, e.g., Clementine– Data interrogation – e.g., ACL, IDEA, MS Access, Excel– Statistical analysis – e.g., SPSS and SAS– Link analysis – I2– Lexis-Nexis– Data conversion utilities (Monarch)– Internet, open-source research– Access to system query tools

Brett M. Baker, PhD, CPA, CISA

16

Forensic Approach

• Start with objectives• Structured brainstorming

– Consider SME conference– Identify indicators of potential fraud and ways to find in data– Process to identify financial risks

• Map out the end-to-end process – Identify systems and key processes– Identify key controls

• Identify and obtain transaction-level data – Record layout– Look at 1000 records before examining all records– ACL, IDEA, and Monarch can read virtually any data format

• Flat files, Delimited files, Dbase files, MS Access, Report files, ….• No file size limits

• Build targeted business rules and run against data• Examine anomalies

Brett M. Baker, PhD, CPA, CISA

17

End-to-End Payment UniverseForensic Audit Approach

AccountingSystems

Federal Reserve SystemCommercial

PayEntitlement

Systems

Disbursing Systems

People Pay Entitlement

Systems

PersonnelSystems

CommercialBank

$$ Treasury Check

ContractingSystems

Data Analysis

CentralContractorRegistry

Brett M. Baker, PhD, CPA, CISA

Unemployment Benefits

EmploymentEmployment

Over-payments

Under-payments

Unemployment Insurance

20

How To Apply In Your Organization

• Decide What You Are Looking For• Assign Personnel With Analytical Skills• Gather Data• Understand Data and Business Rules• Select Detection Method/Tools• Produce and Research Anomalies• Refine the Detection Process• Discover the Irregular/Illegal Transactions• Improve the Business Process

– Automated oversight– Continuous monitoring