Operational Risk [30]

Go to: Summary | Previous | Next   
Bullet points include: Method of moments. Vary parameters to achieve exact fit to successive moments of distribution, e.g. mean, then variance (or sum of squared values?), then … Often relatively convenient but basic MoM may not always return valid (or meaningful) parameter values. Generalised method of moments: achieve 'best' fit given some criterion for 'best'. Maximum likelihood. Identify ‘most likely’ choice of parameters. Methodology explained further in Session 4, including some of its attractive asymptotic features

Contents | Prev | Next | ERM Lecture Series

Desktop view | Switch to Mobile