Modelling Risk



Modelling Risk

The concept of Model Risk

Risk measurement relies on modelling assumptions. Errors in these assumptions introduce errors in risk measurement. This makes risk measurement vulnerable to model risk, which refers to the uncertainty of the model being used. In practice, model risk is sometimes addressed by comparing the results of different models for an extensive treatment of this idea with applications to many different markets. An alternative approach to model uncertainty is to mix multiple models. This idea is developed from a Bayesian perspective and applied to portfolio selection. More often, if it is considered at all, model risk is investigated by varying model parameters. Importantly, the consider model errors that go beyond parameter sensitivity to consider the effect of changes in the probability law that defines an underlying model. This includes those model errors that are not reflected in parameter perturbations (Ghoshal, 1987).

To work with model errors described by changes in probability laws, there is a need a way to quantify such changes, and for this, the concept of relative entropy is used in the proper manner. Relative entropy offers a non-parametric way to describe the difference between two probability distributions. In Bayesian statistics, the relative entropy between posterior and prior distributions measures the information gained through additional data. In characterizing model error, we interpret relative entropy as a measure of the additional information required to make a perturbed model preferable to a baseline model. Thus, relative entropy becomes a measure of the plausibility of an alternative model. Therefore, all the issues related to Modelling Risk will be discussed in detail.

Business Risk Audit Approach

Business risk-audit theory argues that a lack of focus on client business risk may have a detrimental impact on audit quality. In response, another audit risk view has emerged over the past couple of decades. In this view, auditors begin with a more holistic understanding of the overall client organization and its environment so as to better align audit resources to those areas of higher business risk. This “business risk audit” approach differs from the audit risk approach in that the auditor more fully engages in an understanding of the company's business strategies and barriers to achieving them. Business risk audit “focuses upon the modelling of business risk processes of the client's company, and using this knowledge as the basis for establishing financial statement risk and, accordingly, the main focus of subsequent audit testing.” This process creates an environment in which the “client company's strategy is examined and tied to its business processes” (MacCrimmon & Wehrung, 1990).

In essence, the business risk audit approach requires the auditor to gain a broader level of knowledge in the client's business than in the traditional audit risk model. It is in this “forest,” instead of “tree-by-tree,” view that auditors are able to build a mental model that accounts for a realization that the whole entity may differ from the sum of its constituent parts. By doing so, auditors are able to refine their lenses their ...