Consequently ratings may not be evenly interpolated 47 Illustration of

Consequently ratings may not be evenly interpolated

This preview shows page 46 - 53 out of 111 pages.

Consequently ratings may not be evenly interpolated
Background image
47Illustration of consequence scale (from COSO)Apply the generic design to each risk type in the Risk Taxonomy using specific relevant risk metrics and descriptions.Apply the generic design to each risk type in the Risk Taxonomy using specific relevant risk metrics and descriptions.All “Extreme”represent risk capacity exceededAll represent immaterial
Background image
48Definition of likelihoodLikelihood is the possibility that a given event will occurIt may be expressed by one or more of:Qualitative descriptor (e.g. “frequent,” “likely”, “possible”, “unlikely”, rare)A percent probability (e.g. 90%)A frequency (e.g. once in every “X” years) or return period (every “X” years) Each rating level should correspond to a non-overlapping range of likelihoods (i.e. be mutually exclusive).Likelihoods will vary in magnitude depending on the time-frame considered. The longer the horizon, usually the greater the probability of an event occurring.Usually a one-year horizon is chosen to align to reporting and financial cycles, however, it is good practice to specify clearly.
Background image
49Designing the likelihood scale:- end points and interpolationLowest rating should capture events that are almost certain – if these are of high consequence we should be well aware that they are imminent!At the other end of the scale the likelihood level should be sufficiently remote, but still within the concern for management and other stakeholders, in order to warrant assessment and treatment.Where numeric definitions are used (such as probabilities or frequencies) the intervals defining each rating may be spaced evenly (i.e. linear interpolation) or other method. Note that it is usual for probability and frequency ranges to be defined on an exponential or logarithmic scale.
Background image
50Illustration of likelihood scale:- AS/NZS ISO 31000A return period of 30 years is equivalent to once in 30 years or 1/30 = rounded probability of 3%
Background image
51The Risk Matrix and risk aversionThe advantage of the Risk Matrix is that it shows how likelihood and consequence are mapped to risk levels for all possible pairwise combinations (i.e. a mapping schema)We can therefore look for coherency when designing risk levels (how risk levels change as likelihood and consequences change)We can also see what combinations of likelihood and consequence have been assigned the same risk level‘Skew’
Background image
52How the Risk Matrix includes Risk AppetiteAnalysing a risk’s likelihood and consequence, and therefore where a risk is placed on the risk matrix, does not involve risk appetite – it is an objective opinion. The risk levels that are assigned to each combination of likelihood and consequence expresses the organisation’s appetite for risk and therefore embed risk aversion.
Background image
Image of page 53

You've reached the end of your free preview.

Want to read all 111 pages?

  • Three '16
  • risk principles

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture