Data are a key requirement for risk analysis models, allowing us to estimate model parameters and to describe the randomness or inter-individual variability of the system being modeled.
The estimation of model parameters is the domain of statistics, of which there are two main philosophical approaches: Bayesian and classical, or frequentist. At Epix Analytics we tend to pick the techniques from each approach depending on how closely they fit our problem and how easy they are to implement. This section offers an in-depth description of both methods, together with the Bootstrap – a classical statistics technique that can release you from some of the restrictive assumptions of classical methods, and a comparison between the results of each method, which turn out to be very similar in most situations.
The determination of distributions of randomness or inter-individual variability from data is also discussed in considerable detail, and covers parametric and non-parametric distributions, first and second order distribution fitting, and Bayesian, classical and Bootstrap methods.
’Garbage in, garbage out’ – so we also discuss how you can evaluate the quality of your data in some depth.