The complexity of Monte Carlo simulations is often solved with brute force

MoCaX Intelligence adds unparalleled sharpness

Monte Carlo simulations’ biggest limitation comes from their numerical noise and the lack of speed if we want to get rid of that noise. This is because we usually need to increase the number of scenarios to a huge number if we want to obtain stable results.

When computing risk analytics via Monte Carlo simulations, the pricing step tends to take most of the computational effort. MoCaX refines that.

For example

  • XVA, XVA sensitivities, IMM-CCR:  In Counterparty Risk and XVA analytics frameworks, the pricing step usually takes between 80% and 95% of the time. When calculating this risk for exotic derivatives, the pricing step can easily take in excess of 99%.
  • FRTB: The same computational limitation is faced by Market Risk calculation when full revaluation is needed. In this case we need to re-price the whole portfolio hundreds or thousands of times. Each re-pricing is often very computationally costly.
  • CCAR, Stress Testing: The new requirements in stress-testing is making banks have to re-price their portfolios a large number of times to the extent that the job is most difficult, often close to impossible.

MoCaX Intelligence gives a new edge to all these tasks. This is achieved by accelerating and decreasing the errors in Monte Carlo simulations to a degree previously thought to be impossible.

Do you want to see how this is done?