Security and Defence Analytics Course

Lesson 1 • Data Types and Measurement Scales

Distinguishes nominal, ordinal, interval, and ratio data common in security datasets. Correct scale identification drives appropriate method selection throughout the course.

Lesson 2 • Probability Fundamentals for Security Analysis

Introduces classical, frequentist, and subjective probability relevant to threat assessment. Grounds students in probabilistic reasoning before advanced modelling chapters.

Lesson 3 • Introduction to Defence Data Sources

Surveys primary and secondary data sources used in security research. Students evaluate source reliability and classification constraints before collecting data.

Lesson 4 • Core Descriptive Statistics Review

Covers measures of central tendency, dispersion, and distribution shape. Provides the quantitative baseline required for all subsequent analytical work.

Lesson 1 • Database Design and Data Governance

Introduces relational database structures and metadata standards for defence data. Governance frameworks ensure data integrity, access control, and long-term usability.

Lesson 2 • Operational Data Collection Methods

Examines sensor logs, incident reports, and field observation as data sources. Students match collection method to analytical objective and operational constraints.

Lesson 3 • Data Cleaning and Preprocessing

Addresses missing values, outliers, and inconsistencies in raw defence datasets. Clean data is a prerequisite for reliable statistical inference in all core chapters.

Lesson 4 • Survey and Sampling Design for Security Studies

Covers probability and non-probability sampling strategies suited to defence populations. Proper design minimises bias and supports valid inference in later analysis.

Lesson 1 • Confidence Intervals for Defence Estimates

Constructs and interprets confidence intervals for means, proportions, and differences. Interval estimation quantifies uncertainty in threat and capability assessments.

Lesson 2 • Power Analysis and Sample Size Planning

Teaches effect size, Type I and Type II error trade-offs, and power calculations. Proper planning prevents underpowered studies that waste defence resources.

Lesson 3 • Non-Parametric and Distribution-Free Tests

Introduces rank-based tests for ordinal or non-normal security data. These methods extend inferential capability when parametric assumptions are violated.

Lesson 4 • Parametric Hypothesis Tests

Covers z-tests, t-tests, and ANOVA for comparing defence-relevant group means. Students select the correct test based on sample size, variance, and data structure.

Lesson 1 • Logistic Regression for Binary Outcomes

Models binary security events such as attack occurrence or mission success. Logistic regression is the primary tool for threat probability estimation in defence.

Lesson 2 • Model Validation and Deployment Considerations

Covers train-test splits, cross-validation, and performance metrics for defence models. Validation ensures models remain reliable when applied to new operational data.

Lesson 3 • Simple and Multiple Linear Regression

Develops ordinary least squares regression for continuous security outcomes. Model diagnostics ensure assumptions are met before predictions are used operationally.

Lesson 4 • Regularisation and Variable Selection

Applies ridge, lasso, and elastic net methods to high-dimensional defence datasets. Regularisation prevents overfitting and improves out-of-sample predictive accuracy.

Lesson 1 • Bayesian Risk Estimation

Applies Bayesian updating to refine threat probability estimates as new intelligence arrives. This section extends the probability foundations from Chapter 1 to operational risk.

Lesson 2 • Extreme Value Theory and Tail Risk

Models low-probability, high-consequence security events using extreme value distributions. Tail risk analysis prevents underestimation of catastrophic threat scenarios.

Lesson 3 • Risk Assessment Frameworks and Metrics

Introduces risk as a function of threat, vulnerability, and consequence. Students map qualitative risk frameworks to quantitative statistical measures.

Lesson 4 • Monte Carlo Simulation for Risk Modelling

Uses random sampling to propagate uncertainty through complex risk models. Simulation outputs probability distributions over risk outcomes for decision support.

Lesson 1 • Time-Series Analysis of Security Events

Applies decomposition, autocorrelation, and stationarity tests to incident time series. Temporal patterns support forecasting and resource pre-positioning decisions.

Lesson 2 • Spatial Statistics Fundamentals

Introduces point pattern analysis, spatial autocorrelation, and hotspot detection. Spatial methods reveal geographic clustering of incidents invisible to non-spatial analysis.

Lesson 3 • Geographic Information Systems for Analysts

Covers GIS data layers, coordinate systems, and spatial joins for security mapping. GIS integration enables analysts to overlay statistical outputs on operational maps.

Lesson 4 • Forecasting Security Event Frequencies

Builds short- and medium-term forecasts of incident counts using statistical models. Forecast accuracy metrics guide model selection for operational use.

Lesson 1 • Supervised Classification for Threat Detection

Covers decision trees, random forests, and support vector machines for binary and multi-class threat labelling. Classification performance is evaluated with defence-relevant cost matrices.

Lesson 2 • Anomaly and Intrusion Detection Methods

Applies statistical and ML-based anomaly detection to network and behavioural data. Anomaly detection is the primary quantitative tool for identifying insider threats and cyberattacks.

Lesson 3 • Clustering and Unsupervised Pattern Discovery

Uses k-means, hierarchical clustering, and DBSCAN to segment adversary behaviours and incidents. Unsupervised methods reveal structure in unlabelled defence datasets.

Lesson 4 • Natural Language Processing for Intelligence Text

Extracts quantitative signals from unstructured intelligence reports using NLP techniques. Text analytics expands the data available for statistical modelling beyond structured records.

Lesson 1 • Designing Analytical Dashboards for Operations

Covers key performance indicator selection, layout principles, and real-time data integration. Dashboards translate ongoing statistical monitoring into actionable operational awareness.

Lesson 2 • Communicating Statistical Uncertainty to Leaders

Translates probabilistic language into terms commanders can act on without distortion. Calibrated communication prevents both overconfidence and decision paralysis in leadership.

Lesson 3 • Statistical Graphics for Defence Briefings

Teaches chart selection, annotation, and uncertainty visualisation for command audiences. Effective graphics reduce misinterpretation of statistical findings by non-technical leaders.

Lesson 4 • Writing Analytical Reports and Assessments

Structures written intelligence assessments with clear key judgments and supporting evidence. Report writing standards ensure findings are reproducible and defensible under scrutiny.