DATA4400 · KAPLAN BUSINESS SCHOOL · T1 2026
Data-driven
Forecasting
Using longitudinal data and statistical models to inform business strategy, anticipate change, and make evidence-based decisions.
12 Weeks of instruction, workshops, and applied projects
3 Assessments across group, individual report, and presentation formats
4 Software platforms: Tableau · Orange · Exploratory · Python
Pre-requisite: STAM4000 · Co-requisite: DATA4100 · Credit Points: 4 · Mode: On-campus
Your Instructor 02 / 07
SV
Stephen Vu
Instructor — DATA4400: Data-driven Forecasting

Education

PhD in Recommender Systems Queensland University of Technology (QUT)

Current Position

Postdoctoral Researcher QUT × Department of Agriculture, Water and the Environment
Data Science & Artificial Intelligence

Teaching Experience

4 years of university teaching QUT · Kaplan Business School · Central Queensland University · James Cook University

Subject Overview 03 / 07

Forecasting is how businesses
see ahead

This subject equips you with the tools and frameworks to analyse time series data — data that evolves over time — and extract insights that drive strategic decisions. You will work with real-world datasets, apply industry-standard methods, and communicate findings like a practising Business Analyst.

From exponential smoothing to ARIMA and Meta's Prophet, you will build a portfolio of methods and the judgement to choose between them in context.

  • LO1 Evaluate the role of time series forecasting in business strategy
  • LO2 Apply analytical and statistical skills to solve forecasting problems
  • LO3 Create and communicate business insights through forecasting
  • LO4 Compare time series forecasting methods
12-Week Journey
W1
Forecasting Foundations
W2
Time Series Properties & GenAI
W3
Moving Averages & Correlation
W4
Exponential Smoothing & Metrics
W5
Assessment 1 (Group)
W6
Prophet for Business
W7
ARIMA & SARIMA
W8
VAR & Multivariate Methods
W9
Regression Models · A2 Due
W10
End-to-End Forecasting Solutions
W11
Adaptive Forecasting
W12
Assessment 3 · Presentations
Why This Subject Matters 04 / 07

"Every strategic business decision is a bet on the future. Forecasting is how you make that bet an informed one."

Career Relevance

The most in-demand analytical skill in business

Demand forecasting, financial planning, supply chain optimisation, and risk management all rely on time series thinking. As an MBA(Analytics) graduate, clients will expect you to deliver these capabilities.

Connecting Data to Decisions

Translating numbers into strategic recommendations

This subject is not just about fitting models — it is about understanding what a forecast means for a business: when to invest, when to hedge, and when to act. That translation skill defines senior analysts.

Real-World Methods

Industry tools used by data teams today

You will use Prophet (Meta), SARIMA, and Tableau forecasting — the same tools deployed by companies like Amazon, Telstra, and government analytics teams for live planning decisions.

Cross-Domain Impact

Applicable across every industry vertical

Retail demand, water security, financial markets, healthcare capacity, climate risk — time series analysis is the common language of all data-driven industries. Your domain will dictate the data; the methods remain yours.

Assessment Structure 05 / 07
ASSESSMENT 01 · WEEK 5 Synthetic Data & Strategic Forecasting 30%
Type Group Presentation (8–10 slides)
GenAI Compulsory (Level 3)

Use GenAI to create a realistic synthetic dataset for an assigned company, validate it quantitatively, run a Tableau forecast, and present strategic insights.

ASSESSMENT 02 · WEEK 9 Fitting & Evaluating Time Series Models 30%
Type Individual Report (500 words)
GenAI Prohibited (Level 1)

Forecast water storage levels using Holt-Winters, Prophet, and SARIMA. Compare models, evaluate out-of-sample accuracy, and advise on water management policy.

ASSESSMENT 03 · WEEK 12–13 Applied Forecasting Project 40%
Type Presentation (10%) + Report (30%)
GenAI Optional (Level 2)

Choose a real-world forecasting problem from your own workplace, public data, or financial markets. Present a 3-minute pitch and submit a 1,000-word analytical report.

A2: GenAI is entirely prohibited. All analysis must reflect your own reasoning and computation.
A3: GenAI optional but must be referenced and documented with all prompts in an appendix.
A1: GenAI is compulsory — you are assessed on how well you use and critique it.
Software & Tools 06 / 07
Visualisation & Forecasting

Tableau Public

Interactive dashboards and built-in time series forecasting. Used in Assessment 1 for presenting synthetic data forecasts. Industry standard for communicating analytical outputs to non-technical stakeholders.

Statistical Analytics

Exploratory.io

A no-code interface for statistical analysis and time series modelling. Provides ACF/PACF plots, decompositions, and model diagnostics — ideal for understanding data structure before modelling.

Machine Learning Pipeline

Orange Data Mining

Visual programming environment for machine learning workflows. Allows rapid experimentation with forecasting pipelines without writing code — valuable for business analysts who need results quickly.

Programming & Automation

Python on Google Colab

Notebook-based Python environment for Prophet, SARIMA, and custom forecasting pipelines. Required for residual diagnostics (ACF/PACF) in Assessment 2. No local installation required.

What to Expect & How to Succeed 07 / 07

What you will do each week

  • Attend 3-hour workshops combining theory, live software demos, and applied exercises
  • Work with real or realistic datasets drawn from business and environmental contexts
  • Apply forecasting methods to increasingly complex, open-ended problems
  • Receive structured feedback at key checkpoints (Week 5, 9, 11)
  • Develop the habit of quantifying business impact — not just reporting model accuracy
Teaching Philosophy "Think like a Business Analyst first. The model is a tool — the insight is the product."

What distinguishes strong work

  • Connecting every forecast output to a dollar-quantified or policy-relevant business outcome
  • Comparing models objectively using out-of-sample metrics — not just in-sample fit
  • Visualisations that tell a story, with clear labels, appropriate scales, and analytical commentary
  • At least four referenced academic or industry sources, correctly cited
  • Submitting complete files to the correct portals on time — project management is assessed

Word limits are strictly enforced — marking stops at the 110% threshold. GenAI over-reliance is penalised across all assessments; original analysis and reasoning are required. Late submissions attract a 5% daily penalty.

← → arrow keys