Diploma in Applied Causal Inference (DAGs, IV, DID)
Online postgraduate diploma of 800 hours in applied causal inference (DAGs, IV, DID). 8 modules covering potential outcomes, directed acyclic graphs, confounder adjustment and double machine learning, instrumental variables, difference-in-differences, causality in time series and transparent causal reporting in line with STROBE, ICH E9(R1) and ISO 3534-3, with an assigned technical tutor.
Diploma in Applied Causal Inference (DAGs, IV, DID) — studying from South Africa
South Africa has a growing digital economy, with strong demand for data scientists, analysts and simulation specialists in finance, mining, telecoms and government. Major banks, mining houses and telecommunications operators such as MTN and Vodacom rely on advanced analytics for planning and operations. Universities including the University of Cape Town and Wits contribute research in machine learning and computational modelling. Protection of personal data is governed by national legislation, which shapes how analytics work is carried out.
Fees for South Africa: R 27 900 (South African rand), with instalments and employer sponsorship available.
Relevant bodies in South Africa
Typical roles
Questions from South Africa
Is a SEIUM qualification recognised by any South African professional body for data and AI?
South Africa has no single statutory register for data scientists or AI practitioners, and bodies such as the Computer Society of South Africa and the South African Qualifications Authority set their own standards. SEIUM is a private online institute whose qualifications are not state degrees and are not accredited by a national body. Recognition is decided by employers or the relevant body, so ask them directly before enrolling.
Do South African employers in data and AI value private online qualifications?
Demand for data and analytics skills in South Africa spans banking, insurance, telecoms, retail and government, and employers often assess skills through portfolios, technical interviews and experience. A private online qualification may support your CV, but it does not guarantee a job or salary outcome. Show prospective employers the syllabus and your project work to gauge their interest.
How does the workload fit with a job, and what English level do I need?
The programme is fully online and taught in English, so you need confident reading and writing skills. As it is remote, no student visa is needed to study from home. Master's programmes total 1,900 hours, postgraduate diplomas 800 hours and specialist courses 300 hours, with an assigned technical tutor and a final project. Fees can be paid in instalments, and employer sponsorship may be possible.
Take the controls.
Run a real engineering trade-off from this field. Move the sliders, watch the model react in real time and beat the challenge.
What you will be able to do
Distinguish correlation from causation using the potential outcomes framework.
Build and analyse directed acyclic graphs.
Adjust for confounding with propensity scores, IPW and double machine learning.
Apply instrumental variables with diagnostics of their assumptions.
Evaluate interventions with difference-in-differences and synthetic control.
Analyse causality in time series and dynamic systems.
Communicate causal results with explicit assumptions and robustness checks.
Who it's for
- Data scientists focused on causal questions.
- Impact analysts.
- Process engineers with an analytical profile who need to know what really drives plant results.
- Quantitative researchers.
- Continuous improvement analysts.
- Advanced analytics consultants.
- Engineering or science graduates seeking to enter manufacturing.
What you'll work with
- STROBE guidelines (Observational studies)
- ICH E9(R1) (Estimands and sensitivity analysis)
- ISO 3534-3 (Vocabulary: design of experiments)
- ISO/IEC 42001:2023 (AI management systems)
- ISO 13053-1 (Six Sigma: DMAIC methodology)
- DoWhy
- EconML
- DAGitty
- R with the did and CausalImpact packages
- Python linearmodels
8 modules
1.1Correlation, confounding and causation1.2Simpson's paradox1.3Rubin's potential outcomes model1.4Average treatment effect (ATE and ATT)1.5SUTVA, positivity and exchangeability assumptions1.6Hierarchy of evidence1.7Observational studies in engineering1.8Internal and external validity1.9Well-formulated causal questions1.10Workshop: formulating causal questions on the shop floor
2.1Pearl's structural causal models2.2Chains, forks and colliders2.3Conditional independences and d-separation2.4Back-door criterion2.5Front-door criterion2.6Collider and selection bias2.7The do-operator and do-calculus2.8Building DAGs with experts2.9DAGitty and DoWhy tools2.10Practical exercise: DAG of a heat treatment process
3.1Regression with covariate adjustment3.2Propensity score matching3.3Inverse probability weighting (IPW)3.4Doubly robust estimators3.5Balance checking3.6Overlap and trimming3.7Double machine learning3.8Heterogeneous effects with causal forests3.9Sensitivity analysis for unmeasured confounding3.10Lab: IPW and DML with EconML
4.1Intuition and assumptions of an instrument4.2Relevance, exclusion and independence4.3Two-stage least squares4.4Weak instruments and diagnostics4.5Local average treatment effect (LATE)4.6Instruments in industrial contexts4.7Designs with imperfect randomisation4.8Regression discontinuity as a related case4.9Limitations and criticisms4.10Practical exercise: 2SLS estimation in linearmodels
5.1Before-after design with a control group5.2Parallel trends assumption5.3Estimation with fixed effects5.4Pre-trend tests5.5Staggered adoption and its problems5.6Callaway-Sant'Anna estimators5.7Synthetic control method5.8Interrupted time series5.9Inference with few groups5.10Case study: process change across several plants
6.1Granger causality and its limits6.2Bayesian structural time-series models (CausalImpact)6.3Causal discovery with PC and FCI6.4LiNGAM linear non-Gaussian models6.5Causality in systems with feedback6.6Interventions in controlled processes6.7Integrating physical knowledge6.8Validating discovered graphs6.9Risks of over-interpretation6.10Lab: CausalImpact on energy consumption
7.1Transparent causal reporting7.2Explicit assumptions and robustness checks7.3Refutation with placebos7.4Visualising effects and uncertainty7.5From estimated effect to decision7.6Ethics and responsibility in conclusions7.7Peer review7.8Integration with experimentation7.9Applied cases in industry7.10Workshop: refuting an analysis with DoWhy
8.1Causal question and context8.2Inventory of available data8.3DAG of the problem8.4Identification strategy8.5Effect estimation8.6Robustness checks8.7Interpretation and decision8.8Limitations and external validity8.9Report and presentation8.10Final project: final review with the tutor
How you'll learn
100% online training on the SEIUM campus, with downloadable material, recorded classes and an assigned technical tutor. There are no fixed timetables: you progress at your own pace within each intake, so the programme is compatible with work.
Each module combines theory, guided reading of the original standard and applied exercises. The working software is DoWhy, EconML, DAGitty, R with the did and CausalImpact packages and Python linearmodels, in free, academic or evaluation versions. Practical work is based on real procedures for estimating effects with observational plant data, difference-in-differences analysis with staggered adoption and refutation and sensitivity tests. Assessment is continuous by module plus the final project.
Standards covered: STROBE guidelines, ICH E9(R1), ISO 3534-3, ISO/IEC 42001:2023, ISO 13053-1.
Build it end to end
Your final project follows the real engineering workflow, step by step, reviewed by your tutor.
Where this takes you
Requirements
- Prior technical training or professional experience in the sector. No university degree is required.
- Internet connection and a computer capable of running the programme software.
- English sufficient to read technical standards.
Ready to start?
Send your application today. Our admissions team replies within 72 hours with your study plan and funding options.
Frequently asked questions
The programme totals 800 hours over 6–8 months. You set your own pace within each intake, and your tutor helps you plan the workload around your job.
You work with DoWhy, EconML, DAGitty, R with the did and CausalImpact packages and Python linearmodels, using free, academic or evaluation versions. Your tutor indicates the free alternative wherever one exists.
A SEIUM Institute diploma, awarded on passing the continuous assessment and the final project. It is SEIUM's own qualification, not an official state degree; its recognition depends on the programme completed and the regulations of each country.
Yes. Modules passed are recognised in a higher programme in the same area (postgraduate diploma or master's). Ask about the recognition policy when you enrol.
Prior technical training or professional experience in the sector (no university degree required), a computer able to run the programme software and English sufficient to read technical standards.
Start dates are set by SEIUM's faculty for each intake. Our admissions team will confirm the next available date when you apply.
Related programmes
Diploma in Design and Operation of Industrial A/B Experiments
Online postgraduate diploma of 800 hours in the design and operation of industrial A/B experiments. 8 modules covering controlled experimentation, power and sample size, switchback and…
Diploma in Scaling, Scheduling and Cost in HPC/Cloud
Online postgraduate diploma of 800 hours in scaling, scheduling and cost in HPC/cloud. 8 modules covering HPC and cloud infrastructures, application scalability, job scheduling with Slurm,…
Diploma in Drift Monitoring and Model Health
Online postgraduate diploma of 800 hours in drift monitoring and model health. 8 modules covering model degradation, statistical and sequential drift detection, label-free performance estimation, drift…
More in Data, AI, Simulation & Analytics
All Data, AI, Simulation & Analytics programmes → · How to apply · Fees & financing · Scholarships