June 2007 Capital Adequacy Extension © Copyright 2007, CCRO. All rights reserved. Page 66 of 92 Establishing How it is Related Once the market risk factors have been related (linked) to components of earnings, the next step is to define how the market risks affect earnings. These relationships can be defined as straight linear or more complex algorithms depending on the resources available to conduct such analysis. Earnings components can be held as a fixed constant, treated as a function of market risk factors, or treated as an independent variable. For example, in defining how power volumes are affected by power prices, there is one of several approaches o Assume volumes are not affected by price, keeping volume projections static regardless of power price level (may be appropriate for utilities and other load serving entities) o Assume a direct relationship, where power volumes have a defined relationship with prices (i.e. conducting a regression analysis between prices and volumes, and basing the relationship on the correlation and beta of the regression). o Model volumes independent of prices. 6.5. Scenario Generation During Scenario Generation, the information from Exposure Mapping is used to produce a series of scenarios that will ultimately produce a distribution of earnings relative to the Expected Earnings. It helps to view earnings as portfolios with components whose value changes with changes in market risk factors. There are a number of methods in which to achieve a value distribution of a multifaceted portfolio closed form methods, Monte Carlo techniques, Econometric models, and others. Figure 6.1 displays the process in generating earnings distributions due to market risk factors.