Skip to content

Risk Academy · Programme outline

Python for Risk

Programming for risk analysis, from clean tables to simulated loss distributions and reproducible notebooks.

outline. not open for enrolmentintermediate · 40 planning hours · prerequisites: none

Learning outcomes

  • Clean, join and validate hazard, exposure and loss tables with data frame tools.
  • Compute exceedance probabilities, return periods and average annual loss from an event table.
  • Run and test a Monte Carlo model of annual aggregate loss and report percentiles with sampling error.
  • Package an analysis as a reproducible notebook or script with pinned dependencies and documented assumptions.

Existing course sources: risk-analytics-python.

New course outlines: Python Risk Data Pipelines; Simulation and Monte Carlo in Python.

Modules, labs and assessments

  1. Tables and tidy data
    Topics: data frames, tidy data, joins, validation rules.
    Lab: Clean a synthetic event and exposure table. Assessment: Short applied task.
  2. Probability in code
    Topics: distributions, sampling, seeds, exceedance curves.
    Lab: Build an exceedance curve from an event table. Assessment: Quiz and worked solution.
  3. Monte Carlo loss models
    Topics: frequency and severity, aggregation, convergence, percentiles.
    Lab: Simulate annual aggregate loss. Assessment: Graded model exercise.
  4. Reproducible work
    Topics: environments, tests, notebooks, provenance.
    Lab: Re-run a colleague's analysis from a lockfile. Assessment: Peer review of a notebook.

Capstone and credential

A reproducible notebook that estimates average annual loss and a 1-in-100 loss for a synthetic portfolio, with a methods memo.

Completion earns a completion record and an achievement badge through the NXS Passport. It records that the learner finished the programme's assessed work. It does not certify competence and is not an accreditation.

ماخذات