(c) 2025 Columbia University Press. All Rights reserved.
Pandora’s Risk: Uncertainty at the Core of Finance Page 119 (139 of 304)
Powered by Tizra® |
Pandora’s Risk: Uncertainty at the Core of Finance resources
Extracted Text (may have errors)
10
Risks in Value- at- Risk
Standard Value-at-Risk (VaR) methodology pretends to
superior identifi cation of the risks of large market losses.
Instead, it indulges the common fallacy of extrapolating
from small losses. It needlessly sacrifi ces precision in big-
picture forecasts and reads too much into minor details.
We can do much better by tracking short-tem volatility
and allowing for uncertainty.
Most fi er a much broader range of outcomes than pay- or- nancial assets off
don’t-pay. Estimating all the probabilities directly is cumbersome and im-
precise. Instead we formulate summary statistics like mean and standard
deviation.
Risk analysts worry a lot about big losses, so they oft en supplement
mean and standard deviation with mea sures of (lower) tail risk. Th at is
where Value at Risk (VaR) comes in. Th - e name appeals, as it suggests a defi
nite cap on losses.
In fact, VaR is closer to a fl ere’s nothing defi nite oor than a cap. Th
about it. And if calculated in the standard way, it is grossly inferior to a host
of other mea sures. Using standard VaR to tame fi nancial risk is like using
cigarette fi lters to tame cancer risk.
A Chance Encounter
’Twas a chance encounter that opened my eyes to VaR chance. I was walking
down the street when a man called out of an alley, “Buddy, can you spare a
million dollars?”
