Master of Business Analytics WRAPPED
Get a summary of what I learned in the first year of my Master of Business Analytics program at the University of Technology, Sydney.
In 2022, I faced the challenging decision of transitioning from a career in mechanical engineering to embark on a new path in data science and business analytics. Commencing my master’s program at UTS in early 2023, the past year has been a great learning experience.
Here’s a brief outline of the courses I took in the first semester, along with a short description of the course contents, so that you can start learning on your own if you don’t wish to spend a fortune. Additionally, I will discuss how I approached the courses and the projects I completed.
Semester 1
1. Foundations of Business Analytics
This course goes beyond a general introduction to the fundamentals of business analytics. It provides a step-by-step guide to core concepts such as data understanding, modeling, storytelling, and visualization.
The final assessment involved a real-life business analyst simulation in which I worked on a project for a consulting company specializing in firm and capital market analysis for investors and financial advisors. This project included tasks such as data preprocessing, fixing data types, data cleaning, and identifying and replacing missing values. Additionally, it involved conducting predictive analysis to determine and evaluate key drivers for revenue and costs in multiple industries.
Concepts Learned: Data storytelling, data visualization, data preprocessing
Tool(s) used: MS Excel and KNIME
Click here for the 2024 handbook for the Foundations of Business Analytics course.
2. Studio 1: Foundation
This course emphasized real-world analytics scenarios, requiring us to identify business problems and propose effective solutions. It had a strong focus on economics, covering topics such as consumer demand, company behavior, market structure, market equilibrium, and government policies.
The course comprised three assignments/projects. The first assignment involved analyzing a real-world business case, providing the flexibility to choose an Australian company and conduct a comprehensive analysis considering both micro-economic and macro-economic factors. The second assessment was a business analytics simulation, where we applied data analytics and economic concepts to address challenges for a ride-sharing company. The final assessment explored Australia’s private and public healthcare systems, examining the impact of various government policy interventions.
Concepts Learned: Regression analysis, demand forecasting
Tool(s) used: MS Excel
Click here for the 2024 handbook for the Studio 1: Foundation course.
3. Data Processing Using SAS
This course covers the essentials of utilizing SAS for data analytics. Before enrolling, I was unfamiliar with SAS (Statistical Analysis Software), but it turns out it’s a leading choice for analytics and report writing.
The first assessment involved three distinct datasets, necessitating the application of basic SAS codes to uncover insights. However, the main goal was to become familiar with the syntax and functions within SAS. The final assessment centred around manipulating a raw dataset, offering a choice between the Amazon Sales Dataset and the Climate Risk and Economic Losses Dataset. Each file had multiple sheets for analysis. If you’re interested in similar datasets, you can explore these options:
Climate Risk and Economic Losses | CIA countries Taxonomy
The goal is to conduct exploratory data analysis using the SAS programming language on these datasets.
Concepts Learned: Data exploration, validation and manipulation using SAS
Tool(s) used: SAS Studio (SAS programming language)
Click here for the 2024 handbook for the Data Processing Using SAS course.
4. Database Principles
This course covers the introductory concepts of database design and implementation. It includes learning about designing entity-relationship (ER) models and converting these models into logical data structures.
The assessment for this course comprises multiple lab exams that progressively test your ability to run SQL queries, challenging you to extract and manipulate relational databases at increasing levels of difficulty. The final project involves creating a custom-tailored database based on individual interests and specific criteria. This includes running SQL queries to extract data addressing specific business problems.
To get an overview of my submission, click HERE.
Concepts Learned: Entity Relationship diagram and model, basic and intermediate SQL querying.
Tool(s) used: PostgreSQL, Lucidchart for ERD
Click here for the 2024 handbook for the Database Principles course.
5. Enabling Enterprise Information Systems
This course delivered a comprehensive exploration of information systems and design thinking. It was relatively theory-heavy, covering key topics including:
- Information systems overview and real-world IS challenges
- Information systems within the organisations
- Organizational strategy, competitive advantage, and information systems
- Ethics, privacy, and information security
- E-business and E-commerce
- Approaches to IS development and project management
- Social, cloud and mobile computing
- Intelligent systems
- Design thinking
The final assessment culminated in a capstone group project, tasking us with designing a prototype for an information service to support business systems. In my group, we developed a low-fidelity prototype for a health and well-being app tailored for truck drivers.
Concepts Learned: Design thinking, prototyping, persona empathy map, Lotus blossom diagram
Tool(s) used: Miro, Canva
Click here for the 2024 handbook for the EEIS course.
Semester 2
1. Studio 2: Specialization
This course exemplifies the real-world responsibilities of data and business analysts, immersing students in solving business problems from scratch using the industry-standard CRISP-DM framework.
What stood out to me was the course’s practical approach, especially in dealing with intentionally poor data quality, the flexibility of tools for dataset analysis, and the absence of a predetermined outcome.
In my group project, we tackled a business case aiming to reduce attrition rates at a US-based pharmaceutical company. The project unfolded as follows:
i) Conducting business and data understanding: We conducted industry research, defined research objectives, and explored initial data (including missing values, range, and outlier analysis).
ii) Data preparation: This involved data preprocessing (cleaning, merging, formatting, and feature engineering), data visualization, and pre-model analysis based on visual insights.
iii) Modeling and Evaluation: We selected and implemented machine learning models (logistic regression, decision tree, and random forest). The best predictive model, considering explainability, predictability, and complexity, was used to form profiles addressing the business problem of reducing attrition rates.
The course’s structure, centred around a substantial business problem, allowed us to gradually solve it throughout the semester.
Concepts Learned: CRISP-DM Framework, Building Machine Learning Models, and Implementation.
Tool(s) used: Python (scikit-learn), IBM SPSS, MS Excel (pivot tables, VLOOKUP, Analysis ToolPak).
Click here for the 2024 handbook for the Studio 2: Specialisation course.
2. Data Visualisation and Visual Analytics
This course, as the name suggests, focused on data visualization and employing effective storytelling techniques for stakeholders. It involved three projects based on intriguing datasets, with one particularly enjoyable project centred around the Australian Open.
The primary tool utilized throughout the course was Tableau, which captivated me instantly. Its capabilities impressed me, and having used it, I’ve decided not to revert to any other tool for visualization. Beyond creating basic charts, I gained proficiency in developing intricate visualizations for storytelling, such as Sankey diagrams, parallel coordinates, treemaps, choropleth maps, etc.
Here are links to my Tableau dashboards and storyboards that I made during this semester:
i) Mapping Australia’s Global Trade Dynamics (1988–2022): A Focus on Mineral Fuels
ii) Grand Slam Analytics: Unveiling Trends and Patterns in the Australian Open
Concepts Learned: High-dimensional visualisation, visual analytics.
Tool(s) used: Tableau
Click here for the 2024 handbook for the Data Visualisation and Visual Analytics course.
3. Data Processing Using Python
This Python course, centered on data analytics, provides a foundational overview of Python. Covering essential topics like comparison operators, functions, object-oriented programming, and data visualization using libraries such as Matplotlib, Seaborn, and Plotly, it also briefly introduces machine learning techniques.
The culmination of the course is the final project, which holds significant weight. This project entails conducting exploratory data analysis on a dataset of our choosing. Personally, I focused on a dataset related to data science job salaries worldwide. For a detailed look at my analysis, you can access my Python notebook through this link:
Data Science Job Salaries: Exploratory Data Analysis
Concepts Learned: object-oriented programming and data visualisation using Matplotlib, Seaborn and Plotly.
Tool(s) used: Pyhon (Libraries: NumPy, Pandas, Matplotlib, Seaborn, Plotly, country converter)
4. Bank Lending and Analytics
This finance course focused on credit risk assessment and marked my initial foray into the field. To prepare, I independently studied various accounting topics, a journey that proved to be worthwhile.
The course featured two significant assessments. The first involved conducting a credit risk assessment for a specific bank, evaluating multiple clients based on various attributes. The final assessment posed a considerable challenge, requiring the analysis of loan-level performance data and the construction of probability of default models for Basel III and IFRS 9 compliance. In essence, PD (Probability of Default) models and LGD (Loss Given Default) models were computed using time series data from 2001 to 2015. These calculations were coupled with datasets encompassing income levels and environmental factors to assess their impact on credit risks and validate the constructed models.
For further insights into this project and an overview of its intricacies, feel free to send me a direct message. Unfortunately, due to confidentiality constraints, I am not permitted to disclose the code publicly.
Concepts Learned: Credit risk assessment, PD and LGD modelling, expected credit losses, loan loss provisioning.
Tool(s) used: Python, MS Excel
Click here for the 2024 handbook for the Bank Lending and Analytics course.
5. Data Processing Using R
This R introductory course centered on data analytics covers fundamental topics such as array operations, data manipulation, and the creation of interactive graphs and dashboards and more. The culmination of the course is a final assessment where students are tasked with building an interactive web application using the Shiny package.
It’s noteworthy that starting in 2024, this course may no longer be available and will be replaced by the Data Processing Using Python course, offering double the credit points.
Concepts Learned: Data analysis using R, building interactive web apps using Shiny
Tool(s) used: RStudio, Shiny
📊 Evaluation
Enrolling in a Master of Business Analytics degree at UTS can be a worthwhile investment if you have the financial means and time commitment. However, for self-learners seeking an alternative path, here’s a concise guideline:
Master MS Excel:
- Solve business problems from scratch using MS Excel.
- Apply the CRISP-DM framework to guide your problem-solving process.
- Learn data cleaning, preprocessing, and storytelling through visualization using MS Excel.
Learn Python:
- Master one programming language, preferably Python.
- Familiarize yourself with key libraries like NumPy, Pandas, Matplotlib, and Seaborn for data manipulation, analysis, and visualisation and solve business problems using CRISP-DM.
- Gain exposure to building machine learning models.
Learn SQL:
- Learn basic and advanced queries and build SQL projects. SQL projects will typically showcase your ability to take a dataset and gather insights.
Tableau for Interactive Dashboards:
- Build interactive dashboards using Tableau.
- Utilize datasets preprocessed using MS Excel or Python for visualisation in Tableau.
Virtual Internships:
- Explore virtual internships, like those offered by Forage, to gain practical experience with big companies.
Project Building and Refinement:
- Generate project ideas using ChatGPT.
- Stick to topics of personal interest to stay focused.
- Build and refine your skills through building projects. Create an end-to-end project with all the tools you’ve learned.
