Applying audit data analytics in financial statement audit a case study at AASC

MINISTRY OF EDUCATION AND TRAINING STATE BANK OF VIETNAM Ho Chi Minh University of Banking DUONG THUY HANG APPLYING AUDIT DATA ANALYTICS IN FINANCIAL STATEMENT AUDIT A CASE STUDY AT AASC GRADUATION TH[.]

MINISTRY OF EDUCATION AND TRAINING STATE BANK OF VIETNAM

Ho Chi Minh University of Banking

DUONG THUY HANG

APPLYING AUDIT DATA ANALYTICS IN FINANCIAL STATEMENT AUDIT:

A CASE STUDY AT AASC

MINISTRY OF EDUCATION AND TRAINING STATE BANK OF VIETNAM

Ho Chi Minh University of Banking

DUONG THUY HANG

APPLYING AUDIT DATA ANALYTICS IN FINANCIAL STATEMENT AUDIT:

A CASE STUDY AT AASC

GRADUATION THESIS

MAJOR IN ACCOUNTING

CODE: 7340301

SUPERVISOR DANG DINH TAN (PhD)

Ho Chi Minh City, 2022

ABSTRACT

Big Data (BD) and Data Analytics (DA) are hot topics in many fields today Enterprises with strong development strategies such as economic groups and large companies are interested in implementing and applying them to their operation processes, and the audit field is no exception However, to apply them in practice, it

is necessary to carefully consider and calculate the potential positives and negatives

of BD and DA In the article, the author summarizes some key features of the application of data analysis and big data in auditing financial statements, and points out the benefits and challenges for business auditing when applicable Thereby, the author discusses solutions to reduce challenges when applying DA and BD for auditing firms in modern financial statement audits

Keywords: Audit data analytics, data analysis, financial statement audit

DECLARATION OF AUTHENTICITY



I declare that all materials presented hereinafter are my own work based on

my best knowledge during university and with the support of my supervisor Dang Dinh Tan (PhD) This thesis is guaranteed to be free from plagiarism, in which any cites or quotes are fully acknowledged in the references

At any time, I am responsible for my works if it is shown that I have violated this declaration of authenticity

Ho Chi Minh City, June 2022

Duong Thuy Hang

ACKNOWLEDGEMENTS



First and foremost, I would like to thank my supervisor, Dang Dinh Tan (PhD), for his excellent direction, assistance, and attention to this thesis This would not have been possible without his unending encouragement and support

Second, I would like to express my heartfelt gratitude to all instructors at Banking University of Ho Chi Minh City for their tremendous efforts in giving me a solid foundation of knowledge in both academic and professional life

Last but not least, I would like to express my heartfelt gratitude to my loving family and friends for their unending patience, support, and faith in me that has enabled me to reach this point This would not have been feasible without their constant encouragement

Ho Chi Minh City, June 2022

Duong Thuy Hang

TABLE OF CONTENTS

ABSTRACT i

DECLARATION OF AUTHENTICITY ii

ACKNOWLEDGEMENTS iii

TABLE OF CONTENTS iv

LIS OF ABBREVIATIONS vii

LIST OF TABLES AND FIGURES viii

CHAPTER 1: INTRODUCTION 1

1.1 The necessity of the topic 1

1.2 General, specific objectives and research questions 2

1.3 Research objects and research scopes 3

1.4 The research methodology 4

1.5 Contribution of the research 4

1.6 Structure of the thesis 4

CONCLUSION OF CHAPTER 1 6

CHAPTER 2: PRIOR STUDIES ON THE APPLICATION OF AUDIT ANALYTICS 7

2.1 Foreign 7

2.2 Domestic 11

CONCLUSION OF CHAPTER 2 14

CHAPTER 3: THEORETICAL PERSPECTIVES 15

3.1 Related concepts 15

3.1.1 Concept of audit data analytics 15

3.1.2 Purpose of audit data analytics 18

3.1.3 Auditor‟s approach to developing specific data analysis tools 20

3.1.4 Reliability of audit data analytics 20

3.1.5 Risks associated with the application of audit data analytics 21

3.2 Applying audit data analytics in financial statement audit 22

3.2.1 Applying audit data analytics in the audit planning phase 22

3.2.2 Applying audit data analytics in the audit implementation phase 24 3.2.3 Applying audit data analytics in the audit reporting phase 27

CONCLUSION OF CHAPTER 3 28

CHAPTER 4: RESULTS AND DISCUSSIONS 29

4.1 Apply some new audit analytics techniques to two client cases of this audit firm accrual 2021 29

4.2 Interviewing results on the advantages, opportunities and challenges of applying those new audit analytics techniques in the context of the audit firm39 4.2.1 The advantages 39

4.2.2 The opportunities and challenges 41

CONCLUSION OF CHAPTER 4 44

CHAPTER 5: CONCLUSION AND RECOMMENDATIONS 45

5.1 Conclusion 45

5.1.1 Advantages 45

5.1.2 Challenges 47

5.2 Recommendations 50

5.2.1 Training on data analytics skills for audit staff 50

5.2.2 Support for auditing firms to apply big data and data analytics in professional practice 51 5.2.3 Auditing firms need to consider carefully before investing in big data and data analytics 52 REFERENCE 54 APPENDIX A: The process of implementing a typical contract at AASC Auditing Firm 58 APPENDIX B: Illustrative Data of Company A for the year 2021 64 APPENDIX C: Illustrative Data of Company B for the year 2021 69

LIS OF ABBREVIATIONS Abbreviations Full meaning

Deloitte Deloitte Touche Tohmatsu

ERP Enterprise resource planning

FPT Financing Promoting Technology

KPMG Klynveld Peat Marwick Geordeler

TP Bank Tien Phong Bank

VACPA Vietnam Association of Certified Public Accountants

LIST OF TABLES AND FIGURES

Figure 1: PwC‟s „Halo for Journals‟ 18

Figure 2: Ending G/L account balances as of 2021 compared to 2020 (in million VND) 32

Figure 3: Changes in ending G/L account balances as of 2021 compared to 2020 (%) 33

Figure 4: Monthly Debit amount of A/R in 2020, 2021 (in million VND) 34

Figure 5: Monthly transactions in 2020, 2021 36

Figure 6: Monthly total amount of transactions in 2020, 2021 (in million VND) 36

Figure 7: Outliers of transaction amounts by month in 2021 37

Figure 8: Number of transactions by week day and month in 2021 39

Table 1: The Balance sheet of company A between 2020 and 2021 32

Table 2: Ten largest transactions in July 2021 of Company B 38

CHAPTER 1: INTRODUCTION 1.1 The necessity of the topic

The analysis is a highly effective audit procedure because it takes little time and is low cost, but it can provide evidence of the uniformity and general plausibility of the data, and at the same time, it helps not to get too involved in the transactions Specifically, analytical procedures are used in all three phases of the audit process to gather evidence to conclude the reasonableness or anomaly of the data

Although numerous scholarly articles have been written on the future of DA in accounting, relatively little empirical scholarly research has been conducted addressing the issues presented in this article However, several scholarly articles present specific research questions to be addressed in DA For example, Wang and Cuthbertson (in press) interviewed a practitioner with more than 30 years of experience developing analytical tools for internal and external auditors They identified eight categories of research questions that scientists can address, including the role of DA in risk analysis, what procedures should be performed, the impact of testing 100% of the population, whether external data should be used, the role of DA use by internal auditors, the interpretation of DA results, the consequences of DA use, and whether the profession needs a DA framework Similarly, Leonard Combs, PwC U.S Chief Auditor and Leader of Auditing Services, Methodology & Tools reports: “Data analytics is changing both the way

we conduct our audits and what those audits deliver It allows us to extract and analyze ever-larger data sets Further use of data analytics will allow us to deliver effective audits more efficiently” (PwC 2017)

Current reality shows that many auditors who apply this procedure are stereotyped, rigid, and do not bring into full play the effect of analytical procedures

in finding and detecting fraud and material misstatement in the report financial statement While data analytics can successfully address some assertions, it does not eliminate the necessity for other audit methods to address the risk of a significant

misstatement as a whole Such techniques include tracing back to the underlying source documents, which is required to resolve other accusations (e.g., occurrence and rights and obligations) As a result, the auditor must perform procedures to address the risk of a significant misstatement as a whole

After completing the internship period at AASC Auditing Firm, the researcher has gained knowledge and experience in the financial statement audit process, especially new data analysis techniques Realizing the need to minimize the risk of errors on the financial statement and optimize the efficiency for users, the author chooses the topic name "Applying audit data analytics in financial statements audit:

A case study at AASC" as graduation thesis

1.2 General, specific objectives and research questions

General objectives:

Currently, in Vietnam, this topic is also being interesting and applied by many organizations and units However, there are not many articles or in-depth studies on the potential benefits and difficulties for auditing firms in Vietnam when applying

BD and DA in the audit of financial statements Therefore, in this article, the author synthesizes and analyzes the characteristics and opportunities, challenges of the application of DA and BD in financial statement audit at AASC Auditing Firm with the desire to contribute a more multi-dimensional view of the financial statements Besides, this study provides insight into how firms‟ leadership and engagement partners and managers perceive the prospects and impediments to audit data analytic (ADA) use

Specific objectives:

Research on the project application process through 2 case studies when auditing financial statements at AASC and conducting in-depth interviews with auditors in AASC:

- Review and explore opportunities and challenges of applying DA, thereby drawing lessons from experience and specific perspectives to apply to audit activities at other auditing firms

- Analyze the preliminary balances in all the accounts in the company's

general ledger to identify unusual changes from the previous year

- Use the results of the analysis to decide whether changes were needed in the

planned nature, timing and extent of the following:

 Other risk assessment procedures, focused on particular accounts and related assertions

 Further audit procedures to be performed in response to assessed risks, including tests of controls and substantive procedures

- To obtain understandings about all transactions occurred in 2021

- To find outliers of transaction amounts in 2021

- To identify risk of frauds in 2021

1.3 Research objects and research scopes

The research subject are two manufacturing and trading companies whose financial statements are audited by AASC Auditing Firm

The research scope:

- Cross section dimension: The study is limited to an enterprise audited by AASC and conducted in Ho Chi Minh City

- Time series dimension: a 2-year period from 2020 to 2021 is selected, in which secondary data is collected from financial statements of two manufacturing and trading companies

1.4 The research methodology

The thesis uses observational and descriptive method to collect primary and secondary information at AASC Moreover, the author uses the theoretical research method and empirical method to search, analyze, process information, and describe the statistical methods, and practical experience through real work

Data sources: In this thesis, the author uses two sources of data, which are primary data and secondary data

- Primary data: Those data were collected by interview method during the research period at AASC, in order to identify the advantages, opportunities and challenges of applying new audit analytics techniques in the context of the audit firm

- Secondary data: That information and knowledge collected from internal AASC documents and customer's accounting data (in Excel)

1.5 Contribution of the research

Firstly, the thesis has systematized the theoretical basis necessary for further studies on ADA Especially, many previous findings related to BD and DA in large audit firms were collected to provide a more thorough perspective on this regard Secondly, the thesis has pointed out new data analysis techniques compared to current ones to apply to improve audit quality

Lastly, policy recommendations based on the research results have been proposed to support Vietnamese auditing firms

1.6 Structure of the thesis

Besides the Introduction, List of tables and figures, References and Appendix, the thesis structure includes 5 chapters as follows:

Chapter 1: Introduction, this chapter gives an overview of the motivation for

the research, research objectives, research questions, research scope as well as research objects and new contributions of the study about applying audit data analytics

Chapter 2: Prior studies on the application of audit analytics, in this chapter,

the author summarizes and presents an overview of previous studies in the world and Vietnam related to the application of DA in auditing financial statements

Chapter 3: Literature review, this chapter indicates contents related to the

research process as well as describes the data source and methods to collect data

Chapter 4: Results and discussions, this chapter describes how auditors apply

DA to the financial statement audit process and discusses the research findings through a case study

Chapter 5: Conclusion and recommendations, this final chapter summarizes

research contents, gives conclusions and thence, some recommendations will be suggested to improve the audit effectiveness of the audit process of financial statement at AASC

CONCLUSION OF CHAPTER 1

In the first chapter, the thesis has stated the necessity of the topic as well as specifically defined objectives, objects and scopes of the research In addition, a number of new contributions of the thesis toward the originalities of this study in the context of Vietnamese auditing firms

CHAPTER 2: PRIOR STUDIES ON THE APPLICATION OF

AUDIT ANALYTICS 2.1 Foreign

Data analytics has become increasingly popular in recent years, especially in the field of accounting and auditing Some studies show that the current testing strategy and acceptance is in its fourth revolution Audit 1.0 (Dai and Vasarhelyi, 2016) is commonly known as the old manual audit There were no computers or software during this time, and technology was almost non-existent The only tools available to any auditor were pens, some paper, and a calculator This paper-based audit was time-consuming and inefficient (compared to current capabilities), but there was simply no other alternative Consequently, this process has been around for centuries The first Information technology (IT) audit revolution was created with technological growth, Audit 2.0 (Dai and Vasarhelyi, 2016) Back then, only 15% of auditors used analysis tools like Excel, which are essential today In recent years, BD has gained popularity and been integrated into these systems, marking the third generation of auditing (Dai and Vasarhelyi, 2016) The latest revolution goes even further with the inclusion of non-financial information in the dataset before

DA is performed With the use of the Internet of Things (IoT), Cyber Physical Systems (CPS), and Smart Factories, DA seems to be becoming increasingly automated, processing large amounts of data to identify patterns and anomalies (Dai and Vasarhelyi, 2016; Dagiliené and Kloviené 2019) Several studies show that DA can improve audits and make them more efficient by increasing the speed, quality, and quantity of information processed The well-known big four accounting firms also use DA daily Deloitte announced a 9.4% more income compared to the previous year Deloitte's global director of people and purpose said that one of his strategies is the use of data analytics Ramlukan (2015), the main and global assurance analytics leader from EY, said

“It is a massive leap to go from traditional audit approaches to one that fully seamlessly integrates big data and analytics The transformed audit will expand

beyond sample-based testing to include analysis of the entire population of relevant data”

audit-PwC and KPMG published articles explaining the importance of using DA, its benefits, but also the barriers to integrating it with traditional auditing However,

DA not only offers benefits but also brings with it several issues and difficulties that audit firms need to consider and handle if they wish to apply it (Earley, 2015) Proper understanding and availability of data are one of the first concerns Second, investor and regulator expectations may change Professional judgment and skepticism, which is crucial for each audit, should not be neglected, so they should

be treated with caution when analyzing DA outcomes and results

The DA was already mainly used in the advisory or tax area According to Deloitte (2013), DA can help reduce tax errors or find ways to reduce tax breaks For example, companies collect huge amounts of data about their customers, suppliers, competitors, or their environment, but do not have the necessary knowledge to process and analyze it properly According to 85% of managers surveyed by KPMG (2015), the biggest challenge is finding the best way to analyze the data collected Although the application of DA in auditing is slower than, for example, forensic or advisory investigations (Katz, 2014; Whitehouse, 2014), DA is presented as the future of auditing (Liddy, 2014) and "has the potential to be the most significant change in the way audits are conducted since the introduction of paperless auditing” (Capriotti, 2014)

In September 2019, PwC announced plans to invest $3 billion in training and technology over the next 4 years (Chawala, 2020) The aim is to relieve the examiners so that they have more time for themselves In some demonstrations, they showed how the automated robotic process can scan invoices, automatically enter data and collect documents, which could ultimately lead to reduced working hours (Dignan, 2020)

KPMG announced in December 2019 that they will invest $5 billion in AI technologies and employee training over five years The outputs are mainly based

on its cloud-based technology to facilitate communication with its customers (Chawala, 2020)

“Now, we can audit all the client‟s transactions with smart data analytics and can also pinpoint transactions with high risk from large datasets Having the ability

to detect patterns in clients‟ data bring it‟s sort of value We can share insights with clients that were hidden in the large piles of data before.” (Consultancy.eu, 2019) Eilifsen et al (2020) interviewed the heads of professional practice (Heads) of five international public accounting firms in Norway to understand the status of the ADA in each company and conducted detailed questionnaires from 206 audit partners and/or managers on ADA performance across 109 audits from the 2017 audited financial statements This research has yielded several important benefits First, little empirical research exists about ADA since they are relatively new technologies and gaining access to auditors is challenging (Austin et al 2019) Gepp et al (2018, 110) argue that qualitative, interview-based studies are needed to fill this knowledge gap Their study provides insight into how firms‟ leadership and engagement partners and managers perceive the prospects and impediments to ADA use Second, they document the prevalence and nature of ADA use on current audit engagements Auditors‟ actual use of ADA reflects auditors‟ judgments and decisions about the efficiency and effectiveness of ADA use Thus, they provide evidence on audit areas where the output from ADA are being used as an evidential source (i.e., sufficient and appropriate audit evidence) and where obstacles to their use remain Finally, their research provides a starting point for practitioners and researchers in validating the efficiency and effectiveness of the use (or non-use) of ADA In sum, their study contributes by documenting the current status of auditors‟ use of ADA at the engagement level and by developing an understanding of why ADA use has not yet fulfilled its promised potential

Their interview results and questionnaire responses show the following First, the firms‟ heads of professional practice express significant uncertainty about how the supervisory inspection authorities will evaluate and accept ADA generated audit

evidence As a result, none of the firms have introduced mandatory use of

“advanced” ADA tools While ADA use is high on the firms‟ agenda and there is a global push for ADA to be used on audit engagements, actual use is limited in their sample Additionally, the firms differ in their strategies in how they implement the use of ADA in their organizations from a “wait and see” approach to centralized ADA functions and extensive firm involvement to facilitate ADA use Second, the partners and managers indicated that their knowledge and training with firm available ADA tools was sufficient to permit their use of ADA and their attitudes towards ADA usefulness are more positive for firm audits in general than for the sampled audit engagements Third, more ADA is used on engagements where the client has an integrated ERP/IT-systems Fourth, there is a higher frequency of ADA use on new audit engagements Participants expressed that recent tenders specifically asked about the use of new technology and ADA in audits and that the audit firms promoted ADA use in the tender process Fifth, they identify where ADA is used across the various phases of the audit In the audit planning phase, ADA is used for the overall assessment of the client's operations and performance, identifying and assessing key risks, and mapping of different processes In the substantive testing phase, ADA is used for journal entry testing, calculating sample-size, selection of random samples, and summarizing ledgers In the completion phase of the audit, ADA is most used for reconciliation and control between final accounts and underlying ledgers, analytical procedures, and final review of financial statements However, overall, the use of ADA in each phase of the audit is low Sixth, they find little use of what would be considered advanced ADA (i.e., statistical regressions, clustering techniques, statistical predictive analysis, computerized process-mapping, etc.) The use of BD and text mining is almost non-existent When ADA is used, ADA output is mostly used as supplementary evidence In summary, based on the interviews with the heads of professional practice and the questionnaire results, their results suggest that the use of ADA within the firms is limited and at an early stage of implementation

The study establishes that the actual use of ADA is low at the engagement level, notwithstanding the many arguments of the potential of ADA usage and the auditing firms‟ commitment to transforming the audit into an ADA-driven product, and points to factors that may explain this outcome Walker and Brown-Liburd (2019) present a conceptual framework that comprehensively describes the emergence of the use of ADA by audit professionals through the lens of institutional theory They argue that in the process of incorporating ADA into the audit, audit firms respond to pressure from the external environment by attempting to gain legitimacy of ADA usage within their environment Given external pressure and the legitimation process, the engagement leader acts and decides whether it makes sense to use ADA on the audit They discuss their results from perspectives of institutional theory to better understand the auditors‟ complex decision of ADA usage and the observed limited use of ADA in their sample Institutional theory provides insight into the underlying causes of the problems facing audit team leaders in applying ADA and how these issues can be mitigated They conclude that institutional pressure for firms to apply ADA primarily stem from technological advancements and to some extent from audit clients, particularly new clients Except for promoting externally that they are engaged in the process of transforming to ADA usage and to some extent towards audit clients; legitimization

of ADA usage to other constituencies seems not to be a high priority for the surveyed firms In deciding the use of ADA at the engagement level, the auditor must be confident in the ability of the ADA tools to efficiently and effectively provide sufficient and appropriate evidence Institutionalized conventions suggest that limited use of ADA and the problems they identify with the use of ADA will persist until ADA usage is proved to be superior to the current evidence gathering process and their use is supported by the firms, regulators, and supervisors

2.2 Domestic

In Vietnam, there have been articles referring to this issue, and enterprises, as well as large corporations, have deployed related applications According to the

survey data of the Ministry of Industry and Trade in 2019, 61% of Vietnamese enterprises are still outside of the 4.0 Revolution and 21% of enterprises have just started preparing activities According to statistics from the National Bureau of Science and Technology Information in 2018, 8% of enterprises use advanced technology; 50% of enterprises use medium- and medium-advanced technologies; the remaining 42% of enterprises use outdated technology The benefits that Industry 4.0 brings to businesses are shown in a 2015 PwC study showing that Industry 4.0 will bring businesses in Asia such as increased revenue (39%), increased production efficiency (68%), and cost reduction (57%)

Among the applied enterprises, many large corporations have had strategies and actions to apply technology to their business activities For example, VinGroup invested in building a Big Data Institute (Vingroup Big Data Institute) in 2018 to research key areas in the Big Data industry, and at the same time to research new and valuable technologies with high application, directly applied to products (VinGroup) In early 2020, FPT Corporation successfully implemented the system building and Big Data analysis for TP Bank, this is FPT's first Big Data contract for banks in Vietnam, including key components: Data Lake data warehouse built on top of the open Hortonworks Data Platform (HDP)- stores Big Data, from multiple sources, including raw and unstructured data pools; The Watson Studio Local machine learning model building platform, combined with the optimal IBM Integrated Analytics System (IIAS) device for high-speed data analysis, reduces model training time In the coming time, FPT IS will continue to deploy consulting Big Data Analyst solutions for Maritime banks (MSB), Techcombank, Vietinbank, BIDV, and Credit Information Center (CIC) (according to FPT Information System) shows that enterprises have been and are ready to apply DA and BD solutions in their main business activities And IBM Vietnam said that Big Data and business analytics solutions are becoming the center of IBM's "transformation" Every day, the world economy generates 2.5 Exabyte of data (equivalent to over 625 million DVDs), and many industries with future strategies will apply Big

Data and DA in production activities its business However, this is still new content and needs a lot of research investment In a survey by KPMG (2014) of CFOs and CTOs conducted in 2014, 99% of respondents noted that data and DA play an important role in their lives with their business strategy, and 96% expressed that they could make better use of Big Data in their organization

CONCLUSION OF CHAPTER 2

In this chapter, the author summarizes and presents an overview of previous studies in the world and Vietnam related to the application of DA in auditing financial statements Based on studying related research articles, the author identifies research gaps to determine the direction for research on the topic and make recommendations

CHAPTER 3: THEORETICAL PERSPECTIVES

3.1 Related concepts

3.1.1 Concept of audit data analytics

Data analytics is the process of processing and examining data to uncover useful information and help users make decisions In auditing, descriptive analysis and diagnostic analysis are the two main types of data analysis that are used

Using data analytics helps the audit team improve their understanding of the data and examine the entire portfolio In addition, data visualization helps uncover trends or correlations in the data, allowing the audit team to focus on high-risk areas

For example, for the risk assessment process related to receivables on the balance sheet, if a preliminary analysis is performed, only the growth or decline over the years can be assessed But by applying data analytics, the auditor can look

at receivables along with the age of the debt over the years, which can assess the magnitude of the increased risk involved Therefore, the engagement team can recommend the appropriate procedures for examining the item

Meanwhile, DA is the method of data or information analysis to draw conclusions and facilitate the decision-making process (World Bank Group, 2017)

As a concept, DA primarily encompasses IT functions and applications, from basic business intelligence (BI), reporting, and online analytical processing (OLAP) to multiple modes of advanced analysis used to analyze data In the exam context, DA involves larger and more complex procedures during the exam process This requires the use of sophisticated software or advanced statistical tools and techniques This can include cluster analysis, predictive modeling, data layers, visualizations, and what-if scenarios that enable the use of new strategies to assess large amounts of relevant audit information The use of analysis tools allows auditors to collect information from internal and external sources as proof in various phases of an examination, e.g., during analytical procedures, testing of controls, risk assessment, and statement-related procedures (Tschakert et al., 2016)

Recent discussions in the audit profession have recognized the importance of

DA in audit practices (Vasarhelyi et al., 2015) As stated by Capriotti (2014), it “has the potential to be the most significant shift in how audits are performed since the adoption of paperless audit tools and technologies” There are at least three main benefits of using DA in an audit, as auditors can take advantage of analytical tools and technology (Gray and Debreceny, 2014; McGinty, 2014) First, DA allows auditors to automate transaction testing, and theoretically, 100% of the population audited can be tested (Liddy, 2014) Second, audit quality can be increased by enabling a better understanding of client processes through the identification and analysis of accounting anomalies (BrownLiburd and Vasarhelyi, 2015; Capriotti, 2014; Whitehouse, 2014) Third, using DA can improve fraud detection in an audit (Earley, 2015)

However, implementing DA in revision is not an easy task There are requirements such as the need to understand the current scope and limitations of the auditing profession before imagining the role of more complex analytics and DA in audit practice (Appelbaum et al., 2017a, b) Salijeni et al (2018) indicated that there are several conceptual discussions in the literature about the factors influencing the use of DA in assessment practices For example, Krahel and Titera (2015) discussed the need for specific accounting and auditing standards related to DA that facilitate the approach, analysis, and presentation of data In addition, the application of DA

in practice can be promoted by appropriate standards with guidelines on questions related to the examination of large amounts of data, e.g., data collection, error response, and auditor competencies Conducting DA is some of the inhibiting factors in including the DA in external audits Empirical evidence from interviews with 21 participants conducted by Dagiliene and Klovien _e (2019) found that firm-related or audit clients (such as size, data-driven strategy, and business model) and institutional aspects (such as competition in the audit market, regulatory policies about BD and DA and educational institutions) are important motivating factors for the application of DA exam practice A questionnaire study by Eilifsen et al (2020)

documented that auditor found DA audit tools simple and not complex enough to use in the course of an audit The auditors also recognized that DA can be applied effectively in audit practice when organizations have adequate DA tools, the necessary skills, and the availability of professional support for the use of DA in audit engagements The auditors also strongly emphasized the importance of integrating customer information systems to enable the use of DA in audits Respondents in a study by Salijeni et al (2018) highlighted the challenges of including the DA in audits, including detecting “false positives” resulting from testing 100% of the population, costs associated with excessive auditing, over-reliance on analytical specialists, and insufficient guidance on auditing standards The benefit of DA in assurance is the ability to use non-financial data (NFD) and external data to better inform audit planning (particularly in risk assessment) and more effectively audit those areas that require judgment, such as valuation or going concerned Because auditors can develop models that can predict future events, often referred to as predictive analytics, they can better help their clients make strategic decisions about their business NFD includes data that the company collects internally, such as human resources data, customer data, marketing data, etc that goes beyond the types of financial statements that auditors normally look

at As pointed out by Alles and Gray (2014, p 16), “the vast majority of data in dig data is NFD”

According to Gartner's IT Glossary, the types of data analysis commonly used

in financial statements are descriptive and diagnostic: Descriptive analysis is the study of data or content to answer the question "What happened?" and is often characterized by traditional business intelligence and visualizations such as pie charts, bar charts, line charts, tables or generated narratives Diagnostic analytics is

a form of advanced analytics that examines data or content to answer the question

"Why did this happen?" and is characterized by techniques such as drill-down, data discovery, data mining, and correlations

3.1.2 Purpose of audit data analytics

The data analysis was developed to improve audit quality The quality of the audit does not lie in the tools themselves, although this obviously cannot be achieved without proper tools, in the quality of the analysis and judgments provided The value is not in the transformation of the data (impressive as that may be), but in the audit evidence extracted from the conversations and queries that the

analysis generates PwC example below:

Figure 1: PwC‟s „Halo for Journals‟1

The information on this dashboard can be used to make comparisons to previous years and (possibly) to other companies If the automatic vs manual indicator shows a high level of manual registrations, it could indicate inefficient use

of the system, the complexity of the process, or depending on the circumstances

1 Illustration from Halo reproduced with permission of PwC Halo © 2016 PwC PwC refers to the PwC network and/or one or more of its member firms, each of which is a separate legal entity Please see www.pwc.com/structure for further details Halo is also a registered trademark owned by PwC All rights reserved

They pose a risk of fraud Metrics related to individual users may also indicate unusual activity that warrants further investigation

It seems clear that the following unique characteristics of data analysis, when applied correctly, can significantly improve audit quality:

 the ability to graphically visualize results: data visualization is now a discipline in its own right;

 sophistication, and the breadth of interrogation options;

 ease of use by non-specialists; and

 scale and speed

Interviewees emphasize different elements of the list above when asked about how data analytics contributes to audit quality For some, the sophistication of inquiries generated by high-quality visualizations has resulted in better-quality explanations For others, more comprehensive and accurate analyses facilitated by the sheer speed and volume of processing are more important, but:

“… auditing standards will eventually catch up with us on this but at the moment, they‟re based on the assumption that what matters is how you go about finding needles in haystacks Data analytics has shrunk the haystacks and, in the future, it‟ll be about what you do with those needles when you‟ve found them.” Auditors can navigate much larger external datasets much more quickly than ever before, as the greatest recent advances have been in the interfaces between client and accountant systems, software, and data interfaces that facilitate data extraction These interfaces allow auditors to perform routines not just as substantive procedures as in the past, but earlier during the audit at the risk assessment stage to understand processes and work on controls

Many of the analyzes that are performed are not fundamentally different from those performed in the past but are now more granular and at the same time broader

in scope Part of the content check, for example by running routines that scan large magazines and unexpected users, which in turn facilitates a further investigation

3.1.3 Auditor’s approach to developing specific data analysis tools

Auditors often use some combination of the following development approaches:

 take the tools and consider how they can be used in the audit;

 ask what has been performed before that might be automated and expanded; or

 commission completely new routines

The first approach is probably the most creative and valuable in the long term The second is safer, more focused, more complainant probably less wasteful While the third approach may seem ideal, it is only just becoming possible

3.1.4 Reliability of audit data analytics

 The firm continues to have difficulty determining where the ADAs can

be used as essential methods to detect misstatements due to error or fraud (i.e., going beyond risk identification and assessment) The main problem is the amount of work required to establish the reliability of the required data When an ADA is used as a material procedure This

is a big hurdle

 The engagement team's completeness test of the trial balance could often identify issues for tracking, as journal entries reflected in the subsystems were sometimes not yet reflected in the general ledger Therefore, it can be difficult to allocate journal entries through the system Finally, the information used to control operations may differ from the information used for financial reporting due to disruptions between aspects of the system

 Analysis of transactional data on a disaggregated basis can highlight adjustment postings or transactions, allowing them to appear as outliers

in visualizations It is important to take this into account and normalize the data to remove the effects of such corrections

3.1.5 Risks associated with the application of audit data analytics

Auditors recognize that confidentiality and security are, and always will critically issues They report that they are very sensitive to risk, but that risk must be managed appropriately, particularly to avoid dysfunctional behavior that can occur when security is overused and users are starting to work entirely outside of the system to get their jobs done

All companies have data processing protocols that detail what personal data they collect, why, how they handle and store it, and what they delete and when This

is often agreed with management Teams are encouraged to minimize the amount of personal information they collect But there are certain risks in this area, and as one respondent puts it: “… B2B generally isn‟t a problem but with payday loan companies, there is sometimes a genuine risk

One of the issues auditors should consider is the possible creation of new personal data when analyzing relate to individuals Not all routines applied to personal data necessarily generate new data Data analysis of journal authorship, even where user IDs are used, may involve the creation of new personally identifiable information

Areas that continue to present challenges include:

 data stored in the Cloud;

 regulatory hurdles, especially those relating to the transfer of data and information across borders (bank audits almost always have to be performed locally, for example)

The challenges discussed with participants are grouped around the following three themes:

1 establishing ADAs as a replacement for traditional audit procedures

2 establishing that ADAs are at least as efficient and effective as traditional audit procedures

3 obtaining the support of entity management and the audit committee regarding the use of ADAs

3.2 Applying audit data analytics in financial statement audit

3.2.1 Applying audit data analytics in the audit planning phase

To prepare well for the audit of financial statements, first, the auditor needs to prepare the data for use

a) Data Acquisition:

Since effective and efficient data collection is one of the critical success factors for using data analytics, engagement teams should determine early on whether the quality of the data that business management can provide is sufficient

to support the analytics being used

Engagement teams may not have the IT skills required to extract the relevant data from the entity's systems in the required format or to organize the data extracted by the client's IT staff into a format suitable for use in data analysis suitable is If this is the case, using specialized personnel and standard scripts for data collection and uploading data analysis tools is good practice to ensure accurate data is obtained in a usable format

Examples of situations that may justify some data transformation include the following:

 when the date format of different systems in an organization is different, for example, "yyyy-mm-dd" format in one system and "dd-mm-yyyy" format in another, or

 where it may be necessary to remove trailing zeros in an inventory item code to ensure proper comparison with another data source, which may not contain such leading and trailing zeros

While some data error issues are relatively easy to resolve, the nature of certain issues identified can cast doubt on the quality of the data the auditor intends

to use For example, in cases where certain data fields are unlikely to contain spaces

or null values, the presence of such elements may indicate that controls over the data are not working effectively and consequently the data in which they may be

used may not be appropriate for the test until the company takes appropriate action

to correct the records

b) Considering Relevance and Reliability of Data

SSA 500 Audit Evidence states that the auditor shall consider the relevance and reliability of the statistics for use as audit evidence

 Relevance refers back to the logical connection with, or bearing upon, the motive of the audit method and, wherein appropriate, the statement below consideration

 Reliability of the statistics is stimulated through its supply and its nature, and the situations below which it's miles obtained, which includes the controls over its guidance and preservation wherein relevant

c) Relevance of Data

With the countless opportunities round statistics analytics, the relevance of statistics becomes more and more essential because the statistics being analyzed want to apply to the audit techniques responding to the chance of cloth misstatement

on the declaration degree of the elegance of transaction or account stability below consideration One instance in which relevance is in query is in which the statistics analytics offer exciting insights to control however does now no longer produce adequate audit evidence

d) Reliability of data

The majority of statistics applied in statistics analytics is IPE, and as such, the auditor is needed to assess whether or not the data is adequately dependable for the auditor‟s purposes, inclusive of as vital with inside the situations below:

 Obtaining audit proof proximately to the accuracy and completeness of the IPE; and

 Evaluating whether or not the IPE is adequately particular and targeted for the auditor‟s purposes

In figuring out the method that the auditor can also additionally use in figuring out whether or not facts are satisfactorily reliable, the reason for which facts

analytics is being performed (for example, whether or not as a danger evaluation procedure, check of controls, significant analytical procedure, or check of details) desires to be considered

For example, if information analytics is used to carry out similar audit processes, the volume of trying out the information underlying the similar audit processes could probably be more than whilst trying out the information meant for use inside the information analytics used as danger evaluation processes These processes to check the underlying information could now no longer be distinctive to the technique taken for conventional audit processes as set out in SSA 500 and might include:

 Obtaining audit proof of the accuracy and completeness of the IPE thru processes carried out simultaneously with the real audit processes implemented to the IPE whilst acquiring such audit proof is a quintessential a part of the audit method itself; or

 Testing the controls over the practice and upkeep of the IPE

If the IT widespread controls are ineffective, the auditor must check their effect on the use of statistics analytics inside the audit

3.2.2 Applying audit data analytics in the audit implementation phase

a) Cash Receipts:

 The use of ADA may not be feasible due to the complexity of the way the audited entity's system processes some transactions For example, there can be many variations in how sales transactions are recorded, including cash receipts This results from rebates, returns, "clearing" of various aspects of transactions, and batch payment processing

 The ADA accustomed map the sales-receivables-receipts method as a part of risk assessment wasn't entirely successful The engagement team had issues in coping with the various completely different sources of information concerning cash receipts

b) Fixed Asset Additions

 When executing the ADA for fixed asset additions, the hiring team was unable to obtain a correct correlation between additions, WIP, and cash payments For example, labor costs are an important part of the fixed asset additions for this company; this was not taken into account when designing the ADA In addition, the audit of this company did not include the conduct

of proceedings at an interim date As a result, problems with this ADA were not identified early enough to develop an appropriate solution The team concluded that for this customer, the additional time and effort required to design and execute the ADA far outweighed the benefits of the ADA Therefore, they resorted to traditional testing methods that were easier to carry out They don't plan to use an ADA for property, plant, and equipment

in the next year

c) Purchases, Payables, Payments

 ADAs used to scan various fields in database files related to purchases and payments improved audit quality by allowing better assessment of the overall risk of fraud related to segregation of duties In that respect they were successful But the ADAs:

- did not replace traditional accounts payable procedures

- were not directed at detecting material misstatements in accounts payable

- did not contribute to audit efficiency

 Accounts Payable integrity is an important requirement This could not be addressed through the use of an ADA Traditional methods were used for this

d) Payroll

 The engagement team does not plan to use an ADA in the payroll audit They believe that traditional auditing methods are better suited for payroll auditing and that using ADA is unlikely to provide management with useful information In addition, previous audits have not identified any significant

risks of material misstatement in payroll transactions or vendor purchases The team wants to focus ADA usage on areas of importance, such as the risk

of fraud related to revenue Company-developed ADAs related to sales verification are readily available The engagement team didn't expect to get very interesting insights from using an ADA for payroll

 It's a bit more complex to use two-way and three-way matches in payroll audit because of the complexity of getting benefits to be paid and who gets different payments (i.e., between the payment earned and the payments received from the employee) However, this result was anticipated before the implementation of the ADA

 ADAs are not used to audit payroll as the payroll function is outsourced A report is obtained from the service inspector and tests are performed on ingress and egress

e) Inventory

 In the short term, the engagement team does not intend to use an ADA

to replace other inventory-related processes In the longer term, for example, there is the possibility of using an ADA to compare the costs with the net realizable value and to replace some traditional audit procedures related to inventory valuation However, the potential value

of the audit is limited as the amount of inventory present tends to be minimal due to the use of a "just-in-time" inventory process Inventory time is spent looking at the physical count It would be more useful to

be able to override stock counts However, it is not intended to use ADA about counting or to replace any aspect of it Additionally, inventory ADA provided less information to share with management than purchases, liabilities, and payments ADA This is likely because inventory management is a primary management approach They spend

a lot of time on it, so it's difficult to identify something that management isn't already aware of

 The engagement team wants to use an ADA other than something Excel-based to develop inventory turnover analysis; however, a tool is still needed Using an ADA as part of inventory verification procedures

is difficult For example, an ADA at an interim date cannot reasonably

be applied to this customer because stock levels vary greatly depending

on the season

3.2.3 Applying audit data analytics in the audit reporting phase

The auditor usually seeks an explanation from management as to the possible reasons for the significant differences found in the data analysis In such cases, additional audit evidence is obtained that supports or contradicts the management's statement The procedures used to obtain this audit evidence may depend, for example, on the nature of the audited account balance and the statement made by management Even when the relevant population is disaggregated, a pattern in the differences may indicate that there is a common explanation for those differences, but this is not necessarily the case

It is often not possible to identify factors that explain the precise amount of a significant difference However, the auditor performs the procedures necessary to obtain sufficient appropriate audit evidence to conclude that the amount of the unexplained portion of the significant difference is not indicative of a material misstatement

Based on the information gained during the investigation, the examiner checks whether the data analysis needs to be further refined Suppose the auditor determines that the differences between the amount recognized and the auditor's expectation are the result of factors that should have been included as part of the design of the data analysis In that case, the auditor may, where possible, consider refining the analysis In doing so, it is important that the investigator remains professionally skeptical and avoid the tendency to seek and interpret the results in a way that confirms the data the investigator already has beliefs or hypotheses

CONCLUSION OF CHAPTER 3

Chapter 3 has provided the most general concepts and fundamental theories related to DA With the benefits and obstacles brought by the project when auditing financial statements, Chapter 4 will show more clearly when applying the project at AASC Auditing Firm

CHAPTER 4: RESULTS AND DISCUSSIONS 4.1 Apply some new audit analytics techniques to two client cases of this audit firm accrual 2021

Case 1: Having been granted the Certificate of Business Registration by the

Department of Planning and Investment, Company A officially came into operation

in October 2015 with the business line of wholesale petroleum and related products The company uses Fast Accounting Offline (FAO) accounting software to easily manage transactions arising during the year FAO with 1 system subsystem and 13 business subsystems, fully meet the accounting and tax requirements for businesses Chief accountants can do accounting with complex models such as many subordinate units, many accounting departments, and many types of foreign currencies, and especially always be updated with circulars on accounting and tax

of the Ministry of Finance and General Department of Taxation, prepare financial statements and accounting books by regulations Carry out collection and expenditure, payment related to cash, deposit, or loan in a strict and detailed manner according to the right object, according to the invoice, according to the contract, and according to the charge item Prevent negative spending With just one click, chief accountants can view a quick report of the cash balance in the fund, deposits, and loans at each bank The sales-collection cycle can be done on the software from order creation to delivery and collection, helping to receive full orders, on-time delivery, and timely collection of money The purchase - payment cycle can be done

on the software from ordering to receiving goods and paying suppliers Reports on inventory status, sales orders, support timely ordering, and optimal inventory Debts are tracked in detail according to the payment due date of each invoice, helping to pay on time and build a good relationship with suppliers Allows accountant to choose to calculate inventory prices by business requirements: monthly average, moving average, or first-in, first-out Extremely fast pricing, even with thousands of material listings and a large number of import and export coupons Along with other

diverse features, FAO has greatly supported the accounting operations at Company

A as required by management

Before conducting report analysis, the auditor will ask the chief accountant to output data (including general journal, liabilities, fixed assets, inventory, balance sheet, ) from accounting software that the company uses In order to reveal any unusual changes from previous year, the auditor obtained data of trial balance (in Excel) from the client and performed some data cleaning procedures to convert data into the appropriate format for analytics The following table show the calculation

of changes of all G/L accounts relating to the Balance sheet of the client

(In million VND)

N

Beginning balance (Dec 31, 2020)

Ending balance (Dec 31, 2021)

% Change

85.8 1

3 2111 - Nhà cửa; vật kiến trúc 16115.0 16115.0

0.0 1

4 2112 - Máy móc; thiết bị -1056.0 -1056.0

0.0 1

-6 2131 - Quyền sử dụng đất 19305.0 19305.0

0.0