Business Risk and Simulation Modelling in Practice: Using Excel, VBA and @RISK

CHAPTER 8 Measuring Risk using Statistics of Distributions

This chapter introduces some key terms and core aspects of risk measurement, focusing on principles and concepts that apply both to general risk measurement and to properties of simulation inputs and outputs. We aim to use an intuitive and practical approach; much of the chapter will be supported by visual displays, mostly using @RISK's graphical features.

8.1 Defining Risk More Precisely

Precise measures of risk are necessary for many reasons, not least because a statement such as “one situation is more risky than another” may be either true or false according to the criteria used. For example, the implementation of risk-mitigation actions that require additional investment may increase the average cost, whilst reducing the variability in a project's outcomes.

8.1.1 General Definition

Risk can be described fairly simply as “the possibility of deviation from a desired, expected or planned outcome”. Note that this allows the deviation to be driven either by event-type or discrete risks, as well as by fluctuations due to uncertainty or general variability.

8.1.2 Context-Specific Risk Measurement

Risk often needs to be described at various stages of a process, such as pre- and post-mitigation. More generally, as noted earlier in this text, one usually has a choice of context in which to operate, and a fundamental role of risk assessment is to support this choice.

Thus, risk represents the idea of non-controllability ...