Journal of Risk Model Validation

The use of the triangular approximation for some complicated risk measurement calculations

Nick Georgiopoulos

  • Normal densities are difficult to work in convolutions with other densities.
  • We can approximate the normal density with a triangular distribution. 
  • The approximation allows fast and efficient calculations of risk measurements where normal densities complicate the calculation. 
  • Closed form solutions for the VaR of the compound Poisson with lognormal densities can be extracted with this approximation.
  • This approximation is a promising for future research in risk management/measurement.

We introduce the triangular approximation to the normal distribution in order to extract closed- and semi-closed-form solutions that are useful in risk measurement calculations. In risk measurement models there is usually a normal distribution together with some other distributions in a portfolio of risks. Exceedance probability or value-at-risk (VaR) calculations for these portfolios require simulations. However, with the use of the triangular approximation to the normal density we can have closed-form solutions for risk measurements using actuarial models that include not only insurance risk, such as gamma- and Pareto-distributed losses, but also financial risk. We also approximate the collective risk model under lognormally distributed severities and estimate its VaR. We evaluate the accuracy of the analytic solutions versus the Monte Carlo estimates and, according to our results, the analytic solutions requiring much less computational time are quite good.

To continue reading...

You need to sign in to use this feature. If you don’t have a account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an indvidual account here: