Value-at-risk (VAR)
Standing on the threshold
A 'one distribution fits all' approach is not the best option for op risk models. Carsten Steinhoff and Rainer Baule explain why a tailor-made model is therefore vital to the accuracy of loss distribution models
Cracking VAR with kernels
Value-at-risk analysis has become a key measure of portfolio risk in recent years, but how can we calculate the contribution of some portfolio component? Eduardo Epperlein and Alan Smillie show how kernel estimators can be used to provide a fast,…
Risk contributions from generic user-defined factors
In this article, Attilio Meucci draws on regression analysis to decompose volatility, value-at-risk and expected shortfall into arbitrary combinations or aggregations of risk factors, and presents a simple recipe to implement this approach in practice
VAR versus expected shortfall
John Hull discusses the limitations of VAR and the relative advantages of an alternative measure, expected shortfall
Valid Assumptions Required: advanced volatility measures
In the next article of his VAR series, Brett Humphreys discusses more advanced methods for estimating volatility.
Operational VAR: meaningful means
Making the assumption that the distribution of operational loss severity has finite mean, Klaus Böcker and Jacob Sprittulla suggest a refined version of the analytical operational value-at-risk theorem derived in Böcker & Klüppelberg (2005), which…
Dealing with seller's risk
The risk of trade receivables securitisations comes from both the pool of assets and the seller of the assets. Vivien Brunel develops a model for securitisation exposures that deals with both risks, and analyses in detail the interplay between debtors'…
Valid Assumptions Required: Volatility
Brett Humphreys reviews the assumptions associated with calculating volatility based on historical data.
Valid Assumptions Required: Historical Simulation VaR
Brett Humphreys discusses the assumptions underlying the calculation of a VAR using the historical simulation methodology.
Valid Assumptions Required: examining forward curve assumptions
Brett Humphreys and Eric Raleigh review assumptions about the forward curve and the difference between relative and absolute dates.
Valid Assumptions Required: confidence level and holding period
In the second article of his series, Brett Humphreys examines the assumptions associated with selecting a confidence level and a holding period for a VaR calculation
Valid Assumptions Required: aggregation
In the first article of this series, in which Brett Humphreys questions some of the assumptions and decisions that go into the calculation of value-at-risk, he focuses on portfolio aggregation.
Operational risk - Operational VAR: a closed-form approximation
Klaus Bocker and Claudia Kluppelberg investigate a simple loss-distribution model for operational risk. They show that, when loss data is heavy-tailed (which in practice it is), a simple closed-form approximation for operational value-at-risk (VAR) can…