Document Type

Conference Proceeding


Modelling and Simulation Society of Australia and New Zealand


Faculty of Business and Law


School of Accounting, Finance and Economics




This is an Author's Accepted Manuscript of: Allen, D. E., & Powell, R. J. (2007). Thoughts on VaR and CVaR. Proceedings of International Congress on Modelling and Simulation. (pp. 1843-1850). Christchurch New Zealand. Modelling and Simulation Society of Australia and New Zealand. Available here


Value at Risk (VaR) is an important issue for banks since its adoption as a primary risk metric in the Basel Accords and the requirement that it is calculated on a daily basis. VaR calculates maximum expected losses over a given time period at a given tolerance level. Conditional Value at Risk (CVaR) measures extreme risk. It calculates the risk beyond VaR. Relative industry risk measurement is also very important to Banks in their management of risk, such as for setting risk concentration limits and developing investment and credit policy. This paper examines market Value at Risk (VaR) and Conditional VaR (CVaR) in Australia from an industry perspective using a set of Australian industries. VaR and CVaR are compared between these industries over time, and a variety of metrics are used including diversified and undiversified VaR, as well as parametric and nonparametric CVaR methods. There has been no prior investigation of industry based VaR metrics in Australia to the authors’ knowledge. The relative riskiness of different industry sectors is examined and using diversified VaR, the study finds the highest risk is in the Technology Sectors, whilst the lowest risk is found in the Finance and Utilities Sectors. Composite riskiness is also explored and the existence of correlation between industry risk rankings over time is found to depend on the number of years of data used. There is evidence of rank correlation over time using a 7 year window approach, but not when using 1 year data tranches. This highlights the importance of using both short and long time frames in order to cover different economic cycles as well as consider current conditions. It is important to note that there is found to be no significant difference between diversified and undiversified industry VaR rankings, or between parametric and nonparametric CVaR approaches. This means that bankers can be reasonably confident of the robustness and consistency of these metrics when calculating and applying them, not only for the purposes of Basel compliance, but also for the determination of industry risk.

Access Rights