Introduction to Data Analytics & R
1 – Introduction to AFIN8015
This course introduces students to the core process of data science in finance, designed for those interested in the fast-growing FinTech (Financial Technology) field.
It focuses on building key computational, statistical, and analytical skills needed to perform data-driven financial analysis in modern financial industries.
Key Highlights
- Hands-on learning through real, data-driven examples.
- Practical training in regression, classification, data management, visualisation, and machine learning.
- Implementation of models using industry-standard data and R software.
Who Should Join
Ideal for students interested in financial analytics, data analysis, predictive modelling, and classification techniques.
Learning Approach
You’ll explore how data science methods are applied to financial service data to uncover meaningful insights.
This is a practical, lab-based course, with seminars held in computer labs that emphasise empirical analysis of real market data.
Tools Used
- Personal computers are encouraged for all sessions.
- All computations and analysis will be done using R.
Assessment Structure
| Assessment | Weighting | Due | Details |
|---|---|---|---|
| Financial Data Analysis 1 | 40% | Week 7 | Students will analyse real-world financial datasets using relevant descriptive statistics and visualisation techniques. |
| Financial Data Analysis 2 | 55% | Week 12 | Students will conduct both quantitative and qualitative analysis using data science tools and techniques, and present their findings. |
Financial Analytics-Data Analysis-Predictive Modelling-Classification Techniques.
2 – Data Science : An Introduction
What is Data Science?
- There isn’t one single definition of data science.
- It is a cross-disciplinary field that combines methods from various domains.
According to Wikipedia,
“Data science is an interdisciplinary field that uses scientific methods, processes, algorithms, and systems to extract knowledge and insights from structured and unstructured data.”
It is closely related to data mining, deep learning, and big data.
In simple terms, data science brings together techniques from:
- Data engineering
- Descriptive statistics
- Data mining
- Machine learning
- Predictive analytics
For this course, the focus will be on applying data science to business and finance, specifically through Financial Data Science — using data-driven methods to understand and solve problems in the financial sector.
Key Component of Data Science
Statistics and Mathematics
Provide the foundation for analysing, interpreting, and modelling data.
Domain Expertise
Involves specialised knowledge and practical skills — for example, financial analytical methods used for classification and prediction.
Computer Science
- Programming skills (e.g., R, Python, and other analytical tools)
- Understanding of algorithms and data structures
- Familiarity with data mining and big data technologies such as Hadoop and NoSQL
Data Engineering and Data Visualisation
Focus on building efficient data pipelines and presenting data insights clearly and effectively.
Machine Learning
- Predictive models that forecast outcomes
- Classification methods that categorise data patterns
Analytics Integration
Combines descriptive, predictive, and prescriptive analytics to transform raw data into actionable insights.
| Type of Analytics | Key Questions | Techniques / Methods |
|---|---|---|
| Descriptive Analytics (Business Intelligence) | • What and when did it happen? • How much is impacted and how often does it happen? • What is the problem? | Statistics |
| Predictive Analytics | • What is likely to happen next? • What if these trends continue? • What if? | Data Mining Predictive Modeling Machine Learning Forecasting Simulation |
| Prescriptive Analytics | • What is the best answer? • What is the best outcome given uncertainty? • What are significantly differing and better choices? | Constraint-based Optimization Multiobjective Optimization Global Optimization |
= (Minelli et al., 2012) major components in a big data analytics framework and the questions it attempts to answer.
Financial Data Science is the application of Data Science to generate insights in the Financial Services domain.
Data Science Domains
| FINANCE Asset pricing, trading strategies and more | HEALTHCARE Epidemiology, Insurance, Image recognition | PHARMACEUTICALS Patient Journey, Treatment Pathways |
| GOVERNMENT Climate Change, Public Policy, Security | MANUFACTURING Supply Chain, Equipment maintenance and others | RETAIL Pricing, Discounts, Market Basket Analysis |
| OIL & GAS Drilling, Sensors, Equipment Maintenance | TRANSPORTATION Airline Promotions, Passenger promotions | UTILITIES Smart Meter Grids, Power consumption |
| WEB INDUSTRY Clickthrough Ads, Marketing | INTERNET SECURITY Log monitoring, Alerts, Detecting intrusions | SPACE & SCIENCE High Energy Physics, R&D and much more |
Data Science Domains (Dasgupta et al., 2018)
The Data Science Process
Data science is all about discovering meaningful patterns and insights from data.
This doesn’t happen all at once — it’s a step-by-step and repetitive (iterative) process that helps us move from a question to a useful, data-driven answer.
The data science process usually includes the following stages:
- Understanding the Problem
Before diving into data, it’s important to clearly define what you’re trying to solve. This stage involves identifying goals, understanding the context, and knowing why the problem matters. - Preparing the Data
Data rarely comes ready to use. This step focuses on collecting, cleaning, and organizing the data so it’s accurate, consistent, and ready for analysis. - Developing the Model
Here, we build a model using statistical and machine learning techniques. The model helps us identify patterns or make predictions based on the data. - Testing the Model
Once a model is built, it’s applied to a new dataset to see how well it performs. This helps us understand if the model’s predictions are realistic and reliable in the real world. - Deploying and Maintaining the Model
After testing, the model can be used in actual business or operational settings. It’s also regularly monitored and updated to keep it accurate as new data comes in or situations change.
Data Science Project Roles
| Role | Responsibilities |
|---|---|
| Project Sponsor | Represents the business interests and champions the project. |
| Client | Represents end users’ interests and serves as the domain expert. |
| Data Scientist | Designs and executes the analytical strategy, and communicates findings with the sponsor and client. |
| Data Architect | Manages data storage and organization; may also oversee data collection processes. |
| Operations | Manages technical infrastructure and deploys the final project results. |
Financial Analytics
Financial Analytics refers to the deep exploration and interpretation of financial data that companies create and use to make smarter business decisions. It helps organizations understand their financial health, identify trends, and plan for the future with confidence.
Financial analytics goes beyond looking at basic numbers — it connects data with real business insights. By applying descriptive, predictive, and prescriptive methods, professionals can uncover meaningful patterns that help:
- Estimate and forecast financial risks and returns
- Evaluate asset pricing and cash flow performance
- Support strategic planning and investment decisions
The techniques used in financial analytics range from simple tools like descriptive statistics and data visualisation to advanced approaches such as:
- Statistical modelling for performance analysis
- Machine learning for pattern recognition and prediction
- Big data technologies for handling large, complex financial datasets
- Time-series modelling to forecast cash flows, profitability, risk, and asset returns over time
In short, financial analytics turns raw financial data into actionable insights, empowering businesses to make informed, data-driven financial decisions.
Types of Financial Data
| Type of Data | Description | Examples |
|---|---|---|
| Time Series Data | Data collected or recorded over time at regular intervals. | Annual, monthly, daily, or intraday financial data (e.g., stock prices, exchange rates). |
| Cross-Sectional Data | Data collected at a single point in time across multiple entities. | Information on different companies or countries at a specific date. |
| Panel Data | Combines both time series and cross-sectional features. | Financial performance of several companies tracked monthly over a year. |
| Unstructured Data | Data that doesn’t follow a predefined format or structure. | Text documents, social media posts, news articles, or emails. |