Singapore Management University

Institutional Knowledge at Singapore Management University

Research Collection School Of Accountancy School of Accountancy

12-2022

The effectiveness of using Python programming approach in

teaching financial analytics
Clarence GOH
Singapore Management University, [email protected]

Yuanto KUSNADI
Singapore Management University, [email protected]

Gary PAN
Singapore Management University, [email protected]

Follow this and additional works at: https://ink.library.smu.edu.sg/soa_research

Part of the Accounting Commons, Higher Education Commons, and the Programming Languages and
Compilers Commons

Citation
GOH, Clarence; KUSNADI, Yuanto; and PAN, Gary. The effectiveness of using Python programming
approach in teaching financial analytics. (2022). 1-32.
Available at: https://ink.library.smu.edu.sg/soa_research/2038

This Working Paper is brought to you for free and open access by the School of Accountancy at Institutional
Knowledge at Singapore Management University. It has been accepted for inclusion in Research Collection School
Of Accountancy by an authorized administrator of Institutional Knowledge at Singapore Management University.
For more information, please email [email protected].
 The Effectiveness of Using Python Programming Approach in
Teaching Financial Analytics

Clarence Goh, Yuanto Kusnadi, Gary Pan

Singapore Management University

Abstract

This paper presents the learning method and challenges regarding implementing a Python-
programming approach in teaching financial analytics to graduate accounting students. The
advent of Big Data as well as related applications and technologies have largely changed the
process and practice of accounting. These have led to essential changes in the construction and
teaching contents of accounting education. While there have been several studies examining
how data analytics is embedded in accounting curriculum, majority of the teaching cases in
accounting focus on analysis and communication with Excel as the principal tool, with very
few of them covering the necessary steps prior to data analysis such as extraction and data
preparation. Therefore, there is an opportunity for academics to create new teaching cases to
provide students with comprehensive experiences throughout the data analytics cycle, perhaps
with other much needed tools, such as Python. In this learning exercise, we describe our
experience in developing and implementing a coding approach towards learning financial
analytics for graduate accounting students as part of a data analytics curriculum. We share
techniques on how to teach coding to accounting students with little or no coding experience,
and some of the practical challenges in the classroom. Results from implementing this module
in a graduate accounting program showed that using Python programming to teach financial
analytics to accounting students was effective and innovative.

Keywords: Python programming approach, financial analytics, teaching innovation

1
 I. INTRODUCTION

This paper demonstrates an effective approach to teaching graduate accounting students

financial analytics through practical application of Python programming. The advent of Big

Data as well as related applications and technologies have largely changed the process and

practice of accounting and led to essential changes in the construction and teaching contents

and practice of accounting education. As a result, we saw a need to embed financial analytics

in our accounting curriculum, establishing students” programming skill and applying it to

enhance financial analysis. Our experience showed that using Python codes helped accounting

students to better understand valuation of financial instruments and portfolio analysis, despite

our students’ lack of substantial background in statistical programming.

Financial analytics refers to the interpretation and use of financial data to garner a

robust understanding of business trends and give insight that allows management to make

strategic, operational and investment decisions in a timely manner. For instance, companies

may use market trends, company financial forecasts, and other economic indicators to set

investment strategies.

In our course, students focus on learning financial analytics through Python

programming to make investment decisions. Our students identify what trading instruments are

available, what investment objectives are, and when is the best time to make an investment

move. From an accountancy perspective, Python is most useful when working with data. It can

essentially read any type, both structured and unstructured. It has powerful capabilities in data

importation and manipulation – tasks like merging and recoding – as well as handling large

amounts. Python can be used for efficient automation. For example, if one needs to run the

same analysis on a new set of sales data each week, doing this in Excel would require opening

a different file manually each week and re-entering formulas and other elements needed for the

analysis, or creating complex automation with Power Query. But one could do that same

2
analysis automatically in a language such as Python, writing a simple script that imports the

new data and runs the same analysis each week, outputting the results in whatever format one

would like.

The benefits of financial analytics include setting clear investment direction; making

timely investment decisions; enhancing market forecasting and comparison-based modelling

(Andriosopoulos et al., 2019). According to a financial analytics market research study1, the

global financial analytics market was valued at $7.6 billion in 2020 and is projected to reach

$19.8 billion by 2030. As such, it is essential for accountants to master financial analytics

knowledge and skill. This is important as there is a growing expectation that accountants will

be in the driver’s seat for the transition companies must make as they take the critical role of

analytics to become a data-driven organization. Moreover in 2018, the AACSB added a

requirement that students master data analytics and that technology be incorporated across the

entire accounting curriculum. Data analytics was also added to the CPA Exam in 2019.

To keep pace with digital revolution, the accounting curriculum offered in universities

are being revamped to allow students who complete a bachelor’s degree, to become competent

in programming with Python upon graduation (Polimeni and Burke, 2021). Qasim and Kharbat

(2020) also highlight data analytics ought to be incorporated in the accounting curriculum so

as to prepare accounting graduates for the workplace and enhance their employability.

While there is little doubt that data analytics knowledge and skills are important to an

accounting education, the challenge for most accounting faculty, however, is how to deliver a

financial analytics course with Python programming. Accounting faculty may not be used to

teaching a programming course, which involves knowing what to code, executing their code

1
https://www.alliedmarketresearch.com/financial-analytics-market.

3
and resolving compiler errors, and debugging the code, knowledge and skills varied

significantly from traditional accounting courses.

This paper aims to plug a gap in the accounting education literature by proposing a

coding approach toward learning financial analytics. Such approach goes beyond case study or

excel-based exercises. The technology and data analytics developments affecting the

accounting profession have a profound effect on accounting curricula. Therefore, it is essential

for academics to integrate current and emerging technologies into accounting curriculums in

order to provide students with the appropriate knowledge and skill sets necessary to succeed in

the accounting profession (Zhan et al., 2018; Polimeni, 2021; Dow et al., 2021). To date, there

have been a number of studies in accounting education that have examined how data analytics

are taught in the classroom. For instance, a common theme is teaching blockchain to accounting

students by using a combination of storytelling and scaffolding approach (Stratopoulos, 2020);

offering teaching syllabus and resources (Stern and Reinstein, 2021); and through coding with

R. Another theme includes teaching Robotic Process Automation (RPA) to accounting students

(Zhang and Vasarhelvi, 2022) and in an audit class (Bellinga et al., 2022). Other themes include

teaching Python programming to management accounting students (Schoute, 2019) and

introducing textual analysis into the accounting classroom (Fisher et al., 2022).

While there have been several studies examining how data analytics is embedded in

accounting curriculum, majority of the teaching cases in accounting focus on analysis and

communication with Excel as the principal tool, with few of them cover necessary steps prior

to data analysis such as extraction and data preparation (Raschke and Charron, 2021).

Therefore, there is a need for academics to create new teaching cases to provide students with

comprehensive experiences throughout the data analytics cycle, perhaps with other much

needed tools. As company’s systems become more automated, it will also be important for

accountants to understand the language that builds these programs and how information can be

4
extracted from it. Understanding coding languages like Python will enable accountants to

understand how data were obtained and review code to ensure the population is error-free and

complete. From performing data reconciliation and document review to being able to

understand the output from a programme that does these tasks for them will become required

skills for accounting graduates as they are entering the profession. To our knowledge, our study

is one of the first teaching cases to implement a Python coding approach in a financial analytics

course. Python skills have become increasingly important to support data analytics work. It

helps accountants with the data extraction and wrangling work that can be so time-consuming,

as well as supporting more statistical analysis. Where accountants are working extensively with

large amounts of data, coding is increasingly likely to become part of the job.

To this end, we describe our experience in developing and implementing a coding

approach towards learning financial analytics for graduate accounting students. We share

techniques on how to teach coding to accounting students with little or no coding experience

and some of the practical challenges in the classroom.

The remainder of the paper is organized as follows. In the next section, we discuss the

motivations and challenges in teaching financial analytics with Python coding to graduate

accounting students. We then discuss our approach for teaching the financial analytics module

by highlighting the challenges and including suggestions for course implementation. Finally,

we acknowledge limitations of our learning experience.

II. MOTIVATION AND CHALLENGES IN PYTHON EDUCATION IN

ACCOUNTING

Growing Demand for Python Skills in Accounting

Python is an important programming tool and platform technology that has many

practical applications in the accounting setting. In addition to being able to handle large

5
volumes of data, Python is also capable of reading both structured and unstructured data types,

and has powerful capabilities in data importation, data manipulation, and statistical modelling.

These characteristics have allowed Python to play a key role in tasks such as automation, model

creation, and data visualization (Gillon 2021). Accordingly, Python has been widely adopted

across a range of accounting and finance functions, including in the areas of financial

accounting (York 2018), management accounting (Schoute 2019), forensic accounting (Miller

and Bryce 2019), audit (AICPA 2019; ICAEW 2021), and financial valuation (Yan 2017).

Organizations which have been reported to have adopted Python include accounting firms,

financial institutions, fintech firms, and hedge funds (Amen 2019; Protasiewicz 2020; Taylor

2020; Hilpisch 2022).

Recognizing the important role that Python can play in their organizations, employers

in the accounting and finance industry increasingly require their employees to possess

knowledge of python programming. For example, the Developer Skills Report highlights that

26.4 per cent of employers in the finance services industry surveyed require employees to

possess Python programming skills (HackerRank 2018). In addition, a recent study by the jobs

portal eFinancialCareers finds that Python is the most commonly requested programming

language by employers in the accounting and finance industry, with 1,497 job postings in 2020

(Butcher 2020). Citigroup, the financial services institution, has also been reported to have

started requiring its analysts to attend Python classes as part of their continuing education

programme. In explaining Citigroup’s motivation for proving such classes for its analysts, Lee

Waite, Citigroup’s head of markets and securities services in North America was quoted as

saying that the bank was “moving more quickly into this world” because “an understanding of

coding seems to be valuable (Surane 2018).”

Despite the growing demand for Python skills from the industry, there is evidence of a

relative lack of job-seekers in the accounting and finance industry who possess the relevant

6
skillsets. For example, the previously cited study by eFinancialCareers also finds that for every

banking job requiring Python skills that is posted on the job portal, there are only 14 candidates

with the relevant skillsets listed on their resumes (i.e. there are only 14 candidates with Python

skills vying for each job posted). In contrast, for every banking job requiring skillsets such as

VBA (PHP), there are 44 (77) candidates with the relevant skillsets listed on their resumes

(Butcher 2020). Speaking at the launch of a study examining job postings for finance and

accountancy roles over the past five years, Andrew Bushfield, a Director at recruitment firm

Robert Half, highlighted the growing demand for employees with technology skills, including

in Python, by saying that “demand for specialist digital skills like Python, SAP, ERP,

Structured Query Language (SQL) and Scrum have been growing for more than five years, but

the COVID-19 pandemic caused a real crunch..the talent to fill this demand is in short supply,

which is pushing employers to increase salary offers for skilled professionals by 10 to 20 per

cent (Williamson 2021).”

Given the demand for skillsets in Python in the accounting and finance sector,

universities have a key role to play in developing talent who possess these skillsets. In 2018,

the Association to Advance Collegiate Schools of Business (AACSB) formally adopted new

accreditation standards for business school accounting programs, and included standard A5,

which specifically requires accounting programs to integrate the teaching of current and

emerging technologies across the curriculum (AACSB 2018). In explaining the motivation for

the implementation of standard A5, Stephanie Bryant, the Chief Accreditation Officer for

AACSB, highlighted that AACBS intends “for the accounting curriculum to be current,

relevant, and forward-looking to prepare students to be successful as they enter the accounting

profession (Vien 2018).” Therefore, to the extent that Python programming skills fulfil a

key demand that employers in the accounting and finance industry have (Williamson 2021),

7
incorporating Python training in the accounting curriculum would play a critical role in

enabling accounting programs to meet standard A5 set by AACSB.

Challenges of Teaching Python to Accounting Students

Accounting instructors face several key challenges in incorporating Python in the

accounting curriculum. First, the use of Python in accounting-related courses is a relatively

recent development in many accounting programs. For example, the course which we examine

in this paper was first introduced only in 2018. Accordingly, instructors often face the

challenging task of having to not only design the courses, but to also familiarize themselves

with the course material, without structured support from their institutions (Gainey,

Gunderman, Slayton, and Splenda 2021). Second, while courses in accounting may employ

Python, they are typically not designed as coding classes aimed primarily at imparting coding

skills. Instead, these courses often focus on the application of Python to solve problems within

the accounting/finance context. In this regard, a key challenge facing instructors relates to how

they should approach and integrate both the Python programming and accounting-application

aspects of the course. Our study addresses these two challenges by providing guidance on the

implementation of a course that uses Python in a specific aspect of accounting/finance

application (i.e. the analysis of financial instruments).

Third, given that accounting students’ academic backgrounds, including in the area of

computer literacy, can differ (Larres, Ballantine, and Whittington 2003; Byrne and Flood 2008),

instructors often find themselves having to teach a class where students’ experience with

Python can vary greatly. For example, one survey respondent in a study examining the

experiences of university instructors who teach data-related courses conducted by Gainey et al.

(2021) highlights the even the coding ability of incoming first-year students can differ

according to the high school they attended. In this regard, one challenge that instructors face

8
relates to the need to design their courses to accommodate students with varying levels of prior

exposure to Python programming. Our study addresses this challenge by evaluating the

effectiveness of a Python course where there is substantial variation in students’ prior

knowledge of Python programming.

III. TEACHING PYTHON TO ACCOUNTING STUDENTS

The most important research question that this paper is attempting to answer is what is the

effectiveness of using Python in teaching various concepts related to financial analytics in one

of the core modules titled “Analytics for Financial Instruments” that postgraduate students of

the Singapore Management University learn in the Master of Science in Accounting (Data

Analytics) program, offered by the School of Accountancy. We also describe the

implementation of the Python programming lessons in the classroom setting.

The Master of Science in Accounting Program

The Master of Science in Accounting (Data Analytics) or MSA in short was launched

in the Academic Year 2018/19. It is the first Master’s in Accounting degree program in Asia,

designed for professionals who are interested in developing expertise in applying data

technology into their domain knowledge to solve business problems, especially in the finance

and accounting sectors. The availability of massive amounts of data has created a high demand

for skilled data analytics professionals who can analyse such data and obtain the insight needed

for informed decision-making. This is extremely vital to accounting and financial services

firms as they have been dealing with large amount of information produced by their clients.

Armed with machine learning and data analytics skills, our MSA graduates excel in using data

to streamline processes, optimize resources and forecast future revenues.

The courses offered in the program include: Programming with Data (using Python),

Data Management (using SQL), Applied Statistics for Data Analytics (using R), Visutal

9
Analytics for Accounting (using Tableau and PowerBI), Forecasting and Forensic Analytics

(using R), and Analytics for Financial Instruments and Value Investing (both using Python).

Analytics for Financial Instruments

“Analytics for Financial Instruments” is one of the core modules that postgraduate students

enrolled in the MSA program have to clear and it is usually offered in the last semester (Term

3) of the Academic Year (from middle of April to end of June). Python programming language

is used to teach the various concepts (as elaborated below) and students need to pass one basic

Python course offered in Term 1 of the Academic Year (middle of August to end of October in

the previous year) as one of the pre-requisites. It is also one of the two courses on Python taken

in the last semester which students will learn to apply the Python programming language in

accounting and finance setting (the other course that uses Python is tittle “Analytics for Value

Investing”).

Classroom Implementation

In this course, students will be introduced to the various financial instruments that can be

traded in the capital markets. In particular, the course will focus on stocks, bonds, and

derivatives (such as options and futures) and how statistical tools can be applied to value and

analyze these important instruments. As the course emphasizes on the practical application of

Python programming language, the instructor will usually upload the Python codes on eLearn

(SMU’s learning management system) that students can download and study, at least one week

before each seminar. During the seminar, the instructor will explain the important codes and

execute the codes to illustrate the various concepts. Jupyter Notebook is used as the platform

to modify and execute the Python codes.

10
 For example, in the first and second seminars, students will learn how to extract basic

financial data from public sources (such as Yahoo! Finance) and how these data can be used

by managers to plot candlestick charts, histogram and return graphs, as well as perform basic

statistics (such as mean, standard deviation, median, skewness, kurtosis, and correlation) make

informative investment decisions for their firms. Please refer to the teaching notes (Stage 1 to

3) for the Python codes that are used in the first two seminars.

More importantly, in the subsequent seminars, students will also be taught the application

of financial theories such as modern portfolio optimization, fixed income analytics, Black

Scholes option pricing model, Monte Carlo simulation and machine learning for forecasting of

financial instruments as well as techniques on hedging. Please refer to the teaching notes (Stage

4) for the Python codes that are used in one application (i.e. Monte Carlo Simulation).

This course is particularly suitable for students who wish to work for financial institutions

as it exposes them use the various analytical tools in Python to generate important inferences

and help managers make value-increasing investment decisions. Therefore, students will be

organized into teams at the start of the course, working collaboratively together to work on a

portfolio optimization project. In the middle of the semester, there is also a guest lecture

arranged where a portfolio manager offers various learning tips to students and illustrates the

real life application of Python in executing modern portfolio strategies.

At the end of the course, students are expected to present their project findings and write

an investment proposal report to recommend the appropriate portfolio diversification strategy

for an investor. The codes that they submit will be verified for correctness by the instructor, to

ensure that each project group has delivered presentation, report as well as Python codes that

that are of high quality.

Effectiveness of Python as a Teaching Tool

11
 In order to test the effectiveness of Python as a teaching tool in the course, MSA students

enrolled in “Analytics of Financial Instruments” offered in Term 3 of AY 2021/22 were asked

to complete a pre-course and post-course survey questionnaires. From a total of 46 students,

40 students (representing a very high response rate of 87%) completed both survey forms,

which were distributed via Google Forms.2 Responses were collected and statistical analysis

were performed on the responses to test the research question of this paper.

Table 1 presents the basic demographics of the survey respondents, which they filled up

in the Pre-Course survey form. 25 (62.5%) of the respondents were female students and the

remaining 15 (27.%) were male students. The age distribution revealed that a majority (80%)

of the students enrolled in the course was in the age group of 18 to 30. Moreover, 30 (75%) out

of the 40 respondents were full-time students, with the remaining 10 (25%) being part-time

students.

[Insert Table 1 here]

In terms of background of undergraduate degree, an overwhelming 78% of the respondents

was from accounting, with another 10% from business. 82.5% of the respondents had working

experience of up to 10 years and only 5% of the respondents were matured, experienced

students with more than 15 years of work experience. It is also not surprising to see that 78%

of the respondents were in either audit or accounting and finance. Only 10% of the respondents

did not have any work experience.

In terms of prior Python programming course taken during undergraduate, only 33% of

the respondents had taken at least one course. Considering that the respondents’ undergraduate

background were mostly from accounting and business, only 1 student reported that he/she had

not taken any accounting and finance course before.

2
To incentivize students to participate in the surveys, a S$10 Starbucks voucher was given to every student who
had completed both the Pre and Post-Course surveys.

12
 Five questions in the Pre-Course survey form were related to respondents’ proficiency

level in using Python programming language, before formally taking the course. Similar

questions were asked in the Post-Course survey form, after the respondents had completed the

course. The survey questions were designed using a seven-point Likert scale ranging from very

low (1) to very high (7), in terms of proficiency. The responses were tracked so as to enable a

pre and post course analysis, which we present in Panels A and B Table 2.

The first question asked is related to the students’ proficiency in Python programming

language in general. Since all of them had taken a basic Python course in Term 1 of the AY

2021/22 before taking course, 50% reported that their proficiency was high (mean = 4.48, std

dev = 1.3). This value increased significantly to 83% (mean = 5.48, std dev = 1.11), after taking

the course. Panel C of Table 2 also reports the difference in the mean values pre and post course,

which was highly significant (t-statistic = 4.79) for the first question. Hence, through the

Python codes taught in the various topics, the course had resulted in an increase in students’

proficiency in Python programming language.

[Insert Table 2 here]

More relevant for the course was the various concepts being taught, namely: obtaining

stock market data from Yahoo! Finance, portfolio analysis, valuation of financial instruments

and forecasting using Python. Before students took the course, majority of them reported low

proficiency in all the concepts mentioned above, ranging from 73% for forecasting to 88% for

valuation of financial instruments. The mean scores were low as well, ranging from 1.98 (for

valuation of financial instruments) to 2.25 (obtaining stock market data from Yahoo! Finance).

These findings from the Pre-Course survey gave us the confidence that the students’

proficiency in the main learning objectives of the course was in general low.

After taking the course, the Post-Course survey revealed a starkly different picture. Now,

students mostly report high proficiency in the concepts taught in class using Python.

13
Forecasting had the lowest mean scores of 5.43 (std dev of 1) and 85% of the students reporting

high proficiency level, Meanwhile obtaining stock market data from Yahoo! Finance had the

highest mean scores of 6.28 (std dev of 1.11) with all students reporting high proficiency level.

The pre and post analysis in Panel C reports consistent findings as the difference in the mean

levels for all the questions being positive and statistically significant at least at the 1% level.

Overall, there is a strong evidence to suggest that the use of Python programming language is

effective in helping students understand and apply the concepts being taught to achieve the

learning outcomes of the course.

Apart from the analysis in Table 2, we further ask seven questions in the Post-Course

survey to understand if students agree that the instructions and Python codes taught in class

help them to understand the various concepts taught, if the group project was relevant in

applying the knowledge learnt on the topic of portfolio analysis, and if students would

recommend using Python in the course and other accounting and finance courses. Similarly, a

seven-point Likert scale questionnaire design was used, ranging from 1 (Strongly Disagree) to

7 (Strongly Agree). The results are presented in Table 3.

We observe that the mean scores for all questions in Table 3 are above 6, ranging from

6.08 (writing Python codes helped me understand valuation of financial instruments) to 6.68 (I

would recommend using Python in finance and accounting courses). More relevantly for the

course, an overwhelming 95% of the students recommend continuing using Python in the

course (mean score = 6.68, std dev = 0.66) the group project was relevant in applying students’

knowledge learnt on the topic of portfolio analysis (mean score of 6.55, std dev = 0.93).

[Insert Table 3 here]

The findings from Table 3 complement that of Table 2, which may also explain why

students were able to acquire high proficiency level in the various concepts learnt using Python.

14
The main reasons were due to the detailed instructions and codes provided earlier to the

students, before the instructor covered the various topics (obtaining stock market data, portfolio

analysis, valuation of financial instruments, and forecasting). The experience in the group

project also allowed students to apply the Python codes learnt in the seminars to recommend

portfolio diversification strategy using real financial data. Finally, the experience in the course

also led students to recommend continuing using Python in this course as well as other finance

and accounting courses. Therefore, the objective of using Python programming language in

teaching financial analytics is achieved in that students are now able to acquire high proficiency

in not only Python, but also the various concepts being taught, so that they can be prepared to

take on more challenges after their graduation.

IV. CONCLUSION

The accounting profession is rapidly evolving due to technological innovations such as

data analytics. The role of accountants will need to develop programming skills as data

analytics continues to enhance and drive changing business models. At the same time,

academic accounting departments in universities are revamping their curriculum to ensure that

their programs equip accounting students with relevant programming and data analytics skills

that will prepare them for the workplace of the future. In this learning exercise, we describe

our experience in developing and implementing a coding approach towards learning financial

analytics for graduate accounting students as part of a data analytics curriculum. We share

techniques on how to teach coding to accounting students with little or no coding experience

and some of the practical challenges in the classroom.

We acknowledge several limitations of this study. Specifically, while Python

programming is introduced in a number of different courses in the curriculum, the survey of

teaching effectiveness was only conducted in one course (Analytics of Financial Instruments).

15
This can be expanded to the other courses so that a more holistic picture on the effectiveness

of using Python programming language can be formed. In addition, the survey respondents

were limited to only 40. We aim to increase the respondents size to a more sizeable number in

future runs of the course.

16
 REFERENCES

Amen, S. 2019. Which version of Python should you learn for finance jobs?

eFinancialCareers. Available at:

https://www.efinancialcareers.sg/news/2019/08/learning-python-for-banking-jobs

American Institute of Certified Public Accountants (AICPA). 2019. Upgrade the financial

statement audit using audit data analytics. Available at:

https://us.aicpa.org/content/dam/aicpa/interestareas/frc/assuranceadvisoryservices/do

wnloadabledocuments/ads-instructional-paper-python.pdf

Andriosopoulos, D., Doumpos, M., Pardalos, P., and C. Zopounidis. 2019. Computational

approaches and data analytics in financial services: A literature review. Journal of the

Operational Research Society 70 (10): 1581-1599.

Association to Advance Collegiate Schools of Business (AACSB). 2018. 2018 Eligibility

procedures and accreditation standards for accounting accreditation. Available at:

https://www.aacsb.edu/-/media/documents/accreditation/accounting/standards-and-

tables/2018-accounting-

standards.pdf?la=en&hash=8DCDA6CE3B0CEF6AB82D39CBF53995DA96111196

Bellinga, J., Bosman, T., Höcük, S., Janssen, W., Khzam, A. 2022. Robotic Process

Automation for the Extraction of Audit Information: A Use Case. Current Issues in

Auditing 16 (1): A1-A8.

Bryne, M, and B. Flood. 2008. Examining the relationships among background variables and

academic performance of first year accounting students at an Irish University. Journal

of Accounting Education 26 (4): 202-212.

Butcher, S. 2020. The coding languages that will get you a job in banking, ranked.

eFinancialCareers. Available at:

17
 https://www.efinancialcareers.co.uk/news/finance/the-six-hottest-programming-

languages-to-know-in-banking-technology

Dow, K., Jacknis, N., Watson, M. 2021. A Framework and Resources to Create a Data

Analytics-Infused Accounting Curriculum. Issues in Accounting Education 36 (4):

183-205.

Fisher, I., Hughes, M., Janvrin, D. 2022. Put your Best Text Forward: Introducing Textual

Analysis into the Accounting Classroom. Issues in Accounting Education 37 (1): 141-

195.

Gainey, M., H. Gunderman, E. Slayton, and R. Splenda. 2021. Understanding the practices

and challenges of teaching with data in undergraduate social science courses at

Carnegie Mellon University. Available at:

https://kilthub.cmu.edu/articles/report/Understanding_the_Practices_and_Challenges_

of_Teaching_with_Data_in_Undergraduate_Social_Science_Courses_at_Carnegie_M

ellon_University/16693072

Gillon, K. 2021. Why you should learn to code. Institute of Chartered Accountants of

England and Wales (ICAEW). Available at:

https://www.icaew.com/technical/business/webinars-and-publications/business-and-

management-magazine/BAM-May-2021/Why-you-should-learn-to-code

HackerRank. 2018. 2018 Developer skills report. Available at:

https://www.hackerrank.com/research/developer-skills/2018/

Hilpisch, Y. 2022. Financial theory with Python. Sebastopol, C. A.: O’Reilly Media Inc.

Institute of Chartered Accountants of England and Wales (ICAEW). 2021. How Eurobank

got to grips with data analytics in internal audit. Available at:

https://www.icaew.com/insights/viewpoints-on-the-news/2021/dec-2021/how-

eurobank-got-to-grips-with-data-analytics-in-internal-audit

18
Kaden, S., Lingwall, J., and Shonhiwa, T. 2021. Teaching Blockchain through Coding:

Educating the Future Accounting Professional. Issues in Accounting Education 36 (4):

281-290.

Larres, P., J. Ballantine, and M. Whittington. 2003. Evaluating the validity of self-

assessment: Measuring computer literacy among entry-level undergraduates within

accounting degree programmes at two UK universities. Accounting Education 12 (2):

97-112.

Miller, P., and C. Bryce. 2019. Learning Python for forensics. Birmingham, U.K.: Packt

Publishing Ltd.

Polimeni, R. 2021. Integrating Emerging Accounting Digital Technologies and Analytics into

an Undergraduate Accounting Curriculum – A Case Study. Journal of Emerging

Technologies in Accounting 18 (1): 159-173.

Protasiewicz, J. 2020. How Python is used in finance and fintech. Netguru. Available at:

https://www.netguru.com/blog/python-in-finance

Raschke, R., and Charron, K. 2021. Review of Data Analytic Teaching Cases, have we

Covered Enough? Journal of Emerging Technologies in Accounting 18 (2): 247-255.

Schoute, M. 2019. Teaching Python to management accounting students: An illustration

using support department cost-allocation methods. The Accounting Educators'

Journal 29 (1): 137-161.

Stern, M., and Reinstein, A. 2021. A Blockchain Course for Accounting and Other Business

Students. Journal of Accounting Education 56: 100742.

Stratopoulos, T. 2020. Teaching Blockchain to Accounting Students. Journal of Emerging

Technologies in Accounting 17 (2): 63-74.

19
Surane, J. 2018. Citi wants analysts to add Python to list of languages on resume. Bloomberg.

Available at: https://www.bloomberg.com/news/articles/2018-06-14/citi-wants-

analysts-to-add-python-to-list-of-languages-on-resume#xj4y7vzkg

Taylor, C. 2020. For workers, 'digital upskilling' puts tech trends on fast-forward. Reuters.

Available at: https://www.reuters.com/article/us-world-work-digitalupskilling-

idUSKBN2771C3

Vien, C. 2018. What to know about AACSB accounting standard A5. Journal of

Accountancy. Available at: https://www.journalofaccountancy.com/newsletters/extra-

credit/aacsb-accounting-standard-a5.html

Williamson, R. 2021. Surprising tech skills in demand for accounting and finance. CPA

Australia. Available at: https://intheblack.cpaaustralia.com.au/careers-and-

workplace/tech-skills-in-demand-for-accounting-and-finance

Yan, Y. 2017. Python for finance. Birmingham, U.K.: Packt Publishing Ltd.

York, D. 2018. Accounting-system: A GAAP compliant basic accounting system in Python

and SQLLite. Available at: https://medmatix.github.io/Accounting-System/

Zhan, J., Her, YW., Hu, T., and Du, C. 2018. Integrating Data Analytics into the

Undergraduate Accounting Curriculum. Business Education Innovation Journal 10

(2): 169-178.

Zhang, C., and Vasarhelvi, M. 2022. How to Teach a 14-Week Robotic Process Automation

(RPA) Course for Accounting Students? Issues in Accounting Education 10.2308.

20
 TABLE 1
Demographics of Graduate Students Enrolled in
Analytics for Financial Instruments

Characteristics of Number of Percentage of

Respondents Respondents Respondents

Gender
Female 25 62.50%
Male 15 37.50%

Age
18 to 25 17 42.50%
26 to 30 15 37.50%
31 to 35 4 10.00%
36 to 40 2 5.00%
Above 40 2 5.00%

Student Status
Full time 30 75.00%
Part time 10 25.00%

Undergraduate Degree Background

Accounting 31 77.50%
Business 4 10.00%
Others 5 12.50%

Working Experience
0 to 5 Years 27 67.50%
6 to 10 Years 6 15.00%
11 to 15 Years 4 10.00%
More than 15 Years 2 5.00%

Field of Work
I do not have any work experience 4 10.00%
Audit 13 32.50%
Accounting & Finance 18 45.00%
Others 5 12.50%

Python Programming course(s)

taken during Undergraduate
Have not taken any course 27 67.50%
At least one course 13 32.50%

Accounting & Finance course(s)

taken during Undergraduate
Have not taken 1 2.50%
1 to 2 1 2.50%
3 to 5 4 10.00%
More than 5 34 85.00%

21
 TABLE 2
Proficiency Level in using Python to Achieve the Learning Outcomes, before and after taking Analytics for Financial Instruments

Panel A: Before taking the course

Std
Please rate your proficiency on the following: Mean Median Min Max % Low % Neutral % High
Dev
1. Python programming language in general 4.48 4.50 1.30 2.00 7.00 22.50% 27.50% 50.00%
2. Obtaining stock market data from Yahoo! 2.25 1.00 1.56 1.00 6.00 77.50% 12.50% 10.00%
Finance using Python
3. Portfolio analysis using Python 2.05 1.50 1.26 1.00 5.00 85.00% 10.00% 5.00%
4. Valuation of financial instruments Python 1.98 1.00 1.23 1.00 5.00 87.50% 7.50% 5.00%
5. Forecasting using Python 2.40 1.50 1.68 1.00 7.00 72.50% 12.50% 15.00%
Panel B: After taking the course
Std
Please rate your proficiency on the following: Mean Median Min Max % Low % Neutral % High
Dev
1. Python programming language in general 5.48 5.50 1.11 3.00 7.00 5.00% 12.50% 82.50%
2. Obtaining stock market data from Yahoo! 6.28 6.00 0.78 5.00 7.00 0.00% 0.00% 100.00%
Finance using Python
3. Portfolio analysis using Python 6.10 6.00 0.87 4.00 7.00 0.00% 5.00% 100.00%
4. Valuation of financial instruments Python 5.68 6.00 1.00 4.00 7.00 0.00% 10.00% 100.00%
5. Forecasting using Python 5.43 5.00 1.15 3.00 7.00 5.00% 10.00% 85.00%
Panel C: Difference in means test
Difference in Mean
Please rate your proficiency on the following: t-statistics p-value
(After – Before)
1. Python programming language in general 1.00 4.79 (0.00)
2. Obtaining stock market data from Yahoo! 4.03 14.70 (0.00)
Finance using Python
3. Portfolio analysis using Python 4.05 17.89 (0.00)
4. Valuation of financial instruments Python 3.70 18.21 (0.00)
5. Forecasting using Python 3.05 11.49 (0.00)
A seven-point Likert scare survey is used where 1 = Very Low, 2 = Low, 3 = Somewhat Low, 4 = Neutral. 5 = Somewhat High, 6 = High, and 7 = Very High

22
 TABLE 3
Descriptive Statistics on Usage of Python in Analytics for Financial Instruments (n = 40)

Std % of % %
Questions Mean Median Dev Min Max Agree Neutral Disagr
Please rate your agreement with the following statements:
1. Instructions and codes given in class helped me understand 6.43 7.00 0.75 4.00 7.00 97.50% 2.50% 0%
portfolio analysis
2. Writing Python codes helped me understand 6.25 6.00 0.84 4.00 7.00 97.50% 2.50% 0%
portfolio analysis
3. Instructions and codes given in class helped 6.21 6.00 0.77 4.00 7.00 97.50% 2.50% 0%
me understand valuation of financial instruments
4. Writing Python codes helped me understand 6.08 6.00 0.86 4.00 7.00 97.50% 2.50% 0%
valuation of financial instruments
5. I would recommend continuing using 6.35 7.00 0.89 4.00 7.00 95.00% 5.00% 0%
Python in this course
6. I would recommend including Python in finance 6.68 7.00 0.66 4.00 7.00 97.50% 2.50% 0%
and accounting courses
7. The group project was relevant in applying your 6.55 7.00 0.93 3.00 7.00 95.00% 2.50% 2.5%
knowledge learnt on the topic of portfolio analysis:
A seven-point Likert scare survey is used where 1 = Strongly Disagree, 2 = Disagree, 3 = Somewhat Disagree, 4 = Neutral. 5 = Somewhat Agree, 6 = Agree,
and 7 = Strongly Agree

23
 APPENDIX 1: TEACHING NOTES

The teaching notes below provides the detailed implementation to learning financial analytics

through programming in Python.

Stage 1: Understanding the Python programming language

In the Master of Science in Accounting (Data Analytics) program, students are

expected to have completed a course titled “Programming with Data”, which includes

learning the basics of data analytics in Term 1 of the program. The skill-sets that they are

expected to acquire in that course includes: learning programming fundamental concepts and

constructs, the building blocks of information systems, through the Python programming

language.

Stage 2: Obtaining financial data using Python and visualizing the data

The first skill-set that students will learn in the course titled “Analytics of Financial

Instruments” is to retrieve financial data from online public data sources such as

https://finance.yahoo.com, using Python. One important requirement of accomplishing the

task is to apply the knowledge they have learnt earlier in Stage 1 on “pandas” , which is a

software library written for Python, that is most commonly used for data manipulation and

analysis as well as machine-learning tasks. Pandas are intended to work with many other data

science packages inside the Python ecosystem and is typically included in every Python

distribution, such as Anaconda. It is particularly useful for time series and tables operations.

As financial data is often organized in time-series format, pandas provide a convenient and

powerful way to retrieve the data (such as daily stock prices) and then to manipulate the data

to obtain information such as mean and standard deviation of the stock returns.

The following codes below illustrates an exercise on how to obtain the stock prices of

Microsoft (MSFT), Apple (AAPL), and Amazon (AMZN) from 1 January 2022 to 31

24
December 2022, to store them into a DataFrame (a two-dimensional, size-mutable, and

potentially heterogenous tabular data), and to print a graph containing the movement of the

three stocks between the periods outlined.

The three tasks can be achieved through importing the relevant packages such as pandas,

pandas_datareader, yfinance, numpy, and matplotlib (as the codes below show).

## Exercise 1: Obtain financial data and print the graph of stock prices

import pandas as pd
Import numpy as np
import matplotlib.pyplot as plt
from datetime import datetime
from pandas_datareader import data as pdr
import yfinance as yf

yf.pdr_override()

# Specify the starting and ending dates for the time series
start = datetime(2022, 1, 1)
end = datetime(2022, 12, 31)

# The first argument is the series we want, second and third are the starting and ending dates
msft = pdr.get_data_yahoo(‘MSFT', start, end)
aapl = pdr.get_data_yahoo(‘AAPL', start, end)
amzn = pdr.get_data_yahoo('AMZN', start, end)

# Select only the adjusted closing price data and convert the data into a DataFrame
stocks = pd.DataFrame({‘MSFT’: msft['Adj Close'],
‘AAPL’: aapl['Adj Close'],
'AMZN': amzn['Adj Close’]})

# Plot the respective graphs

stocks.plot(title='Plot of MSFT, AAPL, and AMZN stock prices', figsize=(14, 10), grid=True)

Output 1:

25
The next exercise for students will be to observe the output above and to point out how to

improve the presentation of graphs. The solution is to use two different scales when plotting

the data: one scale used for AMZN and AAPL, and another scale used for MSFT. The codes

would have been taught to them in the earlier course on “Programming with Data”

## Exercise 2: Better presentation of output using different scale

stocks.plot(secondary_y = [‘MSFT, ‘AAPL’], title = 'Plot of MSFT, AAPL and AMZN stock prices’, grid = True)

Output 2:

26
Stage 3: Financial analytics using Python

Having learnt how to obtain stock price data and visualize them, Stage 3 is for

students to learn how to perform financial analytics on the data that has been obtained using

Python. In Exercise 2 below, students will learn how to present the relevant summary

statistics for the 3 stocks extracted from Yahoo! Finance earlier. They are also expected to

separately present the mean, standard deviation skewness and kurtosis of the distribution of

the 3 stocks, rounding the output to 4 decimal places. The tasks can be achieved by important

scipy and using the various methods such as describe(), aggregrate(), kurtosis(), and skew()

to present the relevant statistics required.

## Exercise 3: Present summary statistics of the three stocks

import scipy
from scipy.stats import skew, kurtosis

# Presenting the summary statistics, rounding to 4 decimal places

stats = round(stocks.describe(), 4)
print(stats)
print('\n')

# Presenting the mean and standard deviation only

aggregate = round(stocks.aggregate([np.mean, np.std]), 4)
print(aggregate)
print('\n')

# Presenting the skewness and kurtosis

kurtosis = round(stocks.kurtosis(),4)
skew = round(stocks.skew(),4)
print('Skewness of the stocks:\n', skew)
print('\n')
print('Kurtosis of the stocks:\n', kurtosis)

27
 print(stats)
print('\n')

# Presenting the mean and std dev only

aggregate = round(stocks.aggregate([py.mean, py.std]), 4)
print(aggregate)
print('\n')

# Presenting the skewness and kurtosis

kurtosis = round(stocks.kurtosis(),4)
skew = round(stocks.skew(),4)

Output 3:
print('Skewness
print('\n')
of the stocks:\n', skew)

print('Kurtosis of the stocks:\n', kurtosis)

MSFT AAPL AMZN

count 252.0000 252.0000 252.0000
mean 267.8030 154.4699 126.2600
std 25.3036 12.8957 23.9935
min 213.6479 126.0400 81.8200
25% 246.4271 144.4749 108.9047
50% 264.7404 153.8273 122.8850
75% 287.7927 165.2807 146.2599
max 333.1979 180.9597 170.4045

MSFT AAPL AMZN

mean 267.8030 154.4699 126.2600
std 25.3036 12.8957 23.9935

Skewness of the stocks:

MSFT 0.2254
AAPL 0.0130
AMZN 0.0466
dtype: float64

Kurtosis of the stocks:

MSFT -0.8385
AAPL -1.0143
AMZN -1.0784
dtype: float64

Next, in Exercise 4 below, students learnt how to compute monthly returns for the 3
localhost:8888/notebooks/Google Drive/My Drive/Python Programs/Topic2/Full Codes_Week 1.ipynb

stocks obtained earlier, using the resample() and pct_change() functions, and to bar charts of

the monthly returns.

## Exercise 4: Compute simple monthly returns and present the bar charts
# Computing percentage change (simple return) and round to 4 d.p.
daily_return = stocks.pct_change()
monthly_return = stocks.resample('M').ffill().pct_change()
02/01/2023, 07:11 Full Codes_Week 1 - Jupyter Noteboo

# Presenting the mean and std dev of daily and monthly returns
Exercise 4c - Presenting daily and monthly returns
daily_stats = round(daily_return.aggregate([np.mean, np.std]), 4)
In print(daily_stats)
[11]: import numpy as np

print('\n')
# Computing percentage change (simple return) and round to 4 d.p.
daily_return = stocks.pct_change()
monthly_return = stocks.resample('M').ffill().pct_change()

monthly_stats = round(monthly_return.aggregate([np.mean,
# Presenting the mean and std dev of daily np.std]),
and 4) monthly returns
print(monthly_stats)
daily_stats = round(daily_return.aggregate([np.mean, np.std]),
print(daily_stats)
4)

print('\n')

# Plotting Bar Chart for monthly

monthly_stats
print(monthly_stats)
returns all 3 stocks
= round(monthly_return.aggregate([np.mean, np.std]), 4)

monthly_return.mean().plot(kind = 'bar', title = 'Plot of Bar Charts for Monthly Returns of MSFT, AAPL, and
# Plotting Bar Chart for monthly returns all 3 stocks
AMZN');
monthly_return.mean().plot(kind = 'bar', title = 'Plot of Bar Charts for Monthly Return

MSFT AAPL AMZN

mean -0.0011 -0.0010 -0.0022
std 0.0222 0.0225 0.0315

Output 4:
MSFT AAPL AMZN
mean -0.0248 -0.0212 -0.0484
std 0.0695 0.0950 0.1263

Exercise 4d - Plotting the

28 cumulative logarithmic returns, h
computing the correlation of the three stocks
 In Exercise 5 below, students further learn how to compute the cumulative

logarithmic returns and the correlations of the logarithmic returns between the 3 stocks as

well as plot the histogram as the scatterplot diagrams. The functions used are the cumprod(),

corr(), and plotting.scatter_matrix().

## Exercise 5: Compute the cumulative logarithmic returns and correlation among the
stocks and present the cumulative compounded log returns, histogram and scatterplot
diagrams.
# Computing the logarithmic return
log_return = np.log(stocks/stocks.shift(1))
log_return = log_return.replace(np.nan, 0)

# Plotting the cumulative logarithmic returns over time; use the cumprod() method to compute
cum_comp_log_return = (1 + log_return).cumprod()

# Plotting the cumulative returns of all stocks

cum_comp_log_return.plot(title = 'Plot of Cumulative Compounded Log Returns over time')
plt.show();

# Plotting the Scattterplot Diagram and Compute and print the correlation matrix, and round up to 4 decimal places
pd.plotting.scatter_matrix(log_return, diagonal = 'hist')

log_return.corr()

Output 5:

29
 Histogram and Scatterplot Diagram

In [13]: log_return.corr()
Correlation of the 3 stocks:
Out[13]:
MSFT AAPL AMZN

MSFT 1.000000 0.821637 0.741830

AAPL 0.821637 1.000000 0.699181

AMZN 0.741830 0.699181 1.000000

Stage
In [ ]: 4: Monte Carlo Simulation using Python

Stage 4 requires applications of financial analytics into advance topics, such as Monte

Carlo Simulation. In Exercise 6 which is illustrated below, an example using the APPL

(Apple, Inc) stock is used whereby we will use the stock prices from January 1 to December

31, 2022 to predict the one year ahead stock prices. The Monte Carlo simulator generates

theoretical future ‘r’ values. Because rate of return is a random number, to model the

movement and determine possible future values, we must use a formula that models the

random movement – called the Brownian motion (applied in the Black-Scholes formula also)

Annual changes in the stock price (i.e. return) = expected growth rate over time + the effect

of constant volatility of people randomly buying and selling the stock over time has on that

expected growth. Therefore, we can model return as follows:

log return = fixed drift rate+random stochastic value

𝑆!
ln # % = 𝛼 + 𝜎 ∗ 𝑍!
𝑆!"#

𝑟 = exp (𝑑𝑟𝑖𝑓𝑡 + 𝑠𝑡𝑜𝑐ℎ𝑎𝑠𝑡𝑖𝑐 𝑡𝑒𝑟𝑚)

30
## Exercise 6: Monte Carlo Simulation of APPL stock price
import pandas as pd
import numpy as np
import datetime as dt
import matplotlib.pyplot as plt
from datetime import datetime
from matplotlib import style
from scipy.stats import norm

from pandas_datareader import data as pdr

import yfinance as yf
yf.pdr_override()

# Specify the starting and ending dates for the time series
start = datetime(2022, 1, 1)
end = datetime(2022, 12, 31)
data = pd.DataFrame()
data['AAPL'] = pdr.get_data_yahoo('AAPL', start, end)['Adj Close']

# Computing logarithmic returns of AAPL

log_returns = np.log(1 + data.pct_change())

# Computing the drift component

u = log_returns.mean()
var = log_returns.var()
drift = u – (0.5 * var)

# Measure the dispersion component

stdev = log_returns.std()

# Creating the random component

Z = norm.ppf(np.random.rand(10,2))

# How far into the future we want to go (365 days) and number of simulations (10)
t_intervals = 365
iterations = 10

# Generating a series of random future returns

daily_returns = np.exp(drift.values + stdev.values * norm.ppf(np.random.rand(t_intervals, iterations)))

# Getting the last traded price of APPL

S0 = data.iloc[-1]

# Create an array of zeroes that follow the same array as the daily returns
price_list = np.zeros_like(daily_returns)
price_list[0] = S0

# Forecasting daily prices for 1 year ahead

for t in range(1, t_intervals):
price_list[t] = price_list[t - 1] * daily_returns[t]

# Plotting the Predicted Stock Prices of AAPL over the next 1 year
plt.figure(figsize=(10,6))
plt.title("Plot of AAPL Predicted Stock Prices")
plt.ylabel("Price (P)")
plt.xlabel("Time (Days)")
plt.plot(price_list)
plt.show()

31
Output 6:

32