Skip to content

Risk Academy · Programme outline

Julia for Risk

High-performance numerical risk computing in Julia, for simulation and optimisation workloads.

outline. not open for enrolmentadvanced · 28 planning hours · prerequisites: Python for Risk

学習成果

  • Write type-stable Julia code and measure its performance against a Python baseline.
  • Implement a vectorised and a multithreaded Monte Carlo loss simulation.
  • Solve a small portfolio or reinsurance optimisation with an open solver interface.
  • Structure a Julia project as a package with tests and a locked environment.

New course outlines: Julia for Numerical Risk Computing.

Modules, labs and assessments

  1. Julia in an afternoon
    Topics: multiple dispatch, types, arrays, packages.
    Lab: Port the Python exceedance curve. Assessment: Short applied task.
  2. Performance
    Topics: type stability, allocations, benchmarking, threads.
    Lab: Benchmark a loss simulation. Assessment: Benchmark report.
  3. Simulation and optimisation
    Topics: random streams, parallel runs, linear and convex programs.
    Lab: Optimise a synthetic reinsurance layer. Assessment: Graded model exercise.
  4. Packages and reproducibility
    Topics: project environments, tests, documentation.
    Lab: Package the model. Assessment: Code review.

Capstone and credential

A tested Julia package that reproduces a Python loss model and reports the speed-up and the differences in results.

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.

出典