Data Analytics and Decision Making

Ali AbdulHussein’s Data Analytics and Decision Making is a four-module University of Windsor course resource. It connects descriptive summaries, predictive models and prescriptive decisions with spreadsheet labs and assignments. Much of the teaching is delivered through online interactive content that is not included in the short repository PDF.

How to study this book

Use the PDF as a course map and access the linked online lectures and laboratory files for the full teaching material. Before each analysis, define the decision, data, target and evaluation criterion. Keep a record of cleaning and modelling choices, and distinguish association, prediction and decision optimisation. The prompts below are original study guidance, not a reconstruction of omitted lectures.

Chapter-wise explanations and study tasks

Module 1 — Introduction to Data Analytics and Decision Making

The published outline introduces careers, data types and repositories, big data, cloud computing and the stages of analytics. It distinguishes descriptive, predictive and prescriptive questions. The PDF principally provides headings and links for these lessons, so it should not be treated as a self-contained technical explanation. Begin with the decision being supported and identify which analytical question could inform it.

Study task: Describe one engineering-management decision and separate the data summary, forecast and recommended action it might require.

Module 2 — Descriptive Analytics

Visualisation and summarisation are paired with spreadsheet labs, recommended statistical reading and an insurance-charge assignment. These activities ask readers to inspect patterns before making stronger claims. A difference between groups in a dataset does not by itself establish a causal effect, and category definitions require care. The interactive explanations and associated data files need to be consulted alongside the PDF.

Study task: Create a data dictionary, a suitable plot and summary statistics for a sample dataset, then state one conclusion the description cannot justify.

Module 3 — Predictive Analytics

The course outline includes regression, multiple and polynomial models, logistic regression, nearest neighbours, validation, resampling and feature selection. Advertising and credit-related labs connect these topics with applications. Prediction must be evaluated on data not improperly used to choose the model, and the outcome type affects the method. The PDF records the sequence and tasks but omits most lecture explanations.

Study task: Specify a target, a baseline and a validation plan for a fictional prediction problem, identifying a possible source of data leakage.

Module 4 — Prescriptive Analytics

Network flow, routing, simulation, investment, decision trees and revenue management connect models with actions under constraints. The PDF lists spreadsheet labs and assignments, including hotel booking, while the main instruction remains online. An optimum is conditional on its objective, constraints and inputs. Check whether the model represents the decision-maker’s real choices before interpreting its recommendation.

Study task: Formulate a small allocation problem with decision variables, an objective, constraints and one uncertain input, then explain a sensitivity check.

Applied assignment

Create an analytics decision brief with a descriptive assessment, a predictive plan and a small prescriptive model. Document the data and assumptions, distinguish what is observed from what is forecast and explain how a changed input could change the recommended action.

Suggested reading plan

Complete the four modules in order, opening the online teaching and lab resources as you reach them. Use the listed assignments as separate practice checkpoints and keep model evaluation distinct from spreadsheet completion. Revisit the decision statement after every module.

Questions for reflection

• Is the task describing, predicting or choosing?
• Does the evaluation test performance on information unavailable during fitting?
• What does the optimisation leave outside its objective or constraints?

Edition and source notes

Based on AbdulHussein’s 2022 University of Windsor resource, with contributions noted in the source from Bahman Naderi and the development team. The guide covers the four module groups using the PDF’s outline, readings, labs and assignment descriptions. Embedded lectures and many interactive elements are absent from that PDF and were not treated as reviewed full-text lessons. These original prompts supplement the course map.

Consult the official source and full text.