Value-at-Risk (VaR)
Definitions, methodologies, and the limitations institutional desks know about
In the early 1990s, J.P. Morgan's chairman Dennis Weatherstone reportedly asked his risk team for a single number summarizing how much the firm could lose on any given day.
The internal product that resulted — RiskMetrics, released as a public technical methodology in October 1994 — was a daily firm-wide risk report estimating a one-day loss threshold at high confidence level, summarized in a table that arrived on Weatherstone's desk by 4:15 PM each afternoon. The 4:15 report became a small landmark in financial risk management: it forced the entire firm's positions through a single statistical machine and produced one summary number that the chairman could read in five seconds. The Value-at-Risk (VaR) framework that emerged from this work became the dominant institutional risk-reporting standard of the 1990s and 2000s, was adopted by the Basel Committee for bank regulatory capital requirements (Basel II 2004; pre-existing 1996 amendment to Basel I introduced VaR for market-risk capital), and survived through the 2008 financial crisis despite spectacular failures of the modeled VaR numbers during stress events.
That is the opening. Finishing a lesson is where it stops being interesting and starts being useful: the full lesson runs to 8 sections and ends with 6 practice questions. A free account is what opens the rest, and the other 255 lessons in the Academy with it. No card.
What this lesson covers
- 1Three methodologies — same metric, very different machinery
- 2Why VaR's empirical failures during stress are systematic, not random
- 3VaR Calculator — Three Methodologies
- 4Parametric VaR in closed form
- 5Parametric one-day VaR for a $10M portfolio at 1.5% daily volatility — three confidence levels
- 6Long-Term Capital Management, August 1998 — VaR's most-studied failure
- 7Where to see this on the platform
- 8Summary