Volume 3 — Valuation and Risk Metrics © Copyright 2002, CCRO. All rights reserved. 19 Because of the above factors, the application of VaR-type metrics to asset activities must be carefully considered. Certain metrics, such as EaR and CFaR, can provide a risk-based assessment. However, the calculation of such metrics requires detailed analyses of the structure and nature of an asset portfolio in order to be meaningful. To more completely capture volumetric and other types of risk in addition to market or price risk, robust sensitivity analysis and stress testing should be performed (see discussion in Section III). The CCRO will endeavor to provide additional guidance on these VaR-type metrics, as well as sensitivity analysis and stress testing, in our next phase of work. Following is a discussion of the most common variation to VaR that is used more for enterprise- wide risk management than the specific management of the trading/marketing activity. In particular, the methodology outlined: • Allows a better understanding of all the drivers that affect a particular earnings stream and how they react to market forces • Can be used to better communicate corporate goals and strategies to the investment community and rating agencies • Enables more informed hedging strategy decisions • Provides a distribution of outcomes, given a certain confidence level, rather than a point estimate that may never actually occur. 4.2 Calculation This methodology is based on the Corporate Metrics publication by Risk Metrics Group and can be used to calculate earnings at risk, cash flow at risk, or similar measures. The technique uses simulation and is more robust than scenario-based techniques for the following reasons: • Market volatility is incorporated into the analysis. • Correlation between market factors is addressed. • The combined effect of all market risk factors and paths that influence a corporation’s earnings is considered rather than the one or two modeled in a particular scenario. • The user determines the confidence level of the particular metric calculated. • This is a flexible technique that allows incorporation of certain market dynamics such as mean reversion and jump diffusion. This methodology is a five-step process that relies heavily on data to perform the simulations. Step 1: Metric Specification The user specifies the variables. • Holding period ― quarterly, annual, or longer • Confidence interval