Value at Risk is not a coherent risk measure because it violates subadditivity.
Mostly true True in substance, wrong in a detail.
Factual: Statements about the present or the past, checked against the record: did they hold up?
What would count as true or false
True if the standard risk-measure literature (e.g. Artzner, Delbaen, Eber & Heath 1999 and later textbooks) shows that VaR is not a coherent risk measure in general, and that the reason is that it fails subadditivity (VaR of a combined portfolio can exceed the sum of the parts' VaRs). Mostly true if VaR is non-coherent as stated, but the sources name subadditivity together with another failed axiom, or note special cases (e.g. elliptical distributions) where VaR is subadditive, and the claim leaves these out without being wrong in general. Mostly false if VaR is non-coherent but the literature says subadditivity is not the reason. False if the literature shows VaR satisfies all coherence axioms in general, or does satisfy subadditivity in general.
Set before the research began.
Evidence (3)
-
E1 Risk Measures, Risk Aggregation and Capital Allocation (IEOR E4602: Quantitative Risk Management lecture notes) Columbia University (lecture notes by Martin Haugh) · 2010
“Value-at-risk is not a coherent risk measure because it fails to be subadditive.”
-
E2 Messages from the academic literature on risk measurement for the trading book (Basel Committee Working Paper No. 19) Basel Committee on Banking Supervision / Bank for International Settlements · 31 January 2011
“VaR is not coherent because it may violate the subadditivity criterion.”
-
E3 Coherent risk measure - Wikipedia Wikipedia
“Value at risk is, however, coherent, under the assumption of elliptically distributed losses (e.g. normally distributed) when the portfolio value is a linear function of the asset prices.”
Reasoning
Columbia lecture notes (E1) and a Basel Committee working paper (E2) both state that VaR is not a coherent risk measure because it fails, or may violate, subadditivity. This matches the claim's general point and its stated reason. However, E3 notes that VaR is coherent when losses are elliptically distributed (e.g. normal) and the portfolio is linear in asset prices. The claim leaves out this special case. The resolution criteria classify that omission as mostly true.
Cite this claim
Nassim Taleb, “Value at Risk is not a coherent risk measure because it violates subadditivity.” (said 2019). Verdict: mostly true. Reliability API, claim c_61162122384b, method 0.1.0+28da144b, dataset release v1.1.2. https://epistacklabs.com/claims/c_61162122384b
This claim is in dataset release v1.1.2, with the same verdict.
Reply to this claim
More factual claims by Nassim Taleb
-
Mixed
“My point was that bike lanes contribute to IMPROVEMENT in well-being and happiness WITHOUT generating economic growth.”
-
Resolved true
“The entire article misses the point that Spitz has been bullish past 3 years.”
-
Mixed
“It is worth noting that binary options on financial”
-
Mostly true
“they are banned by most corresponding regulators, as they are”
-
Mostly false
“We know that economic forecasts work no better than astrology”
Verdicts are made automatically; how verdicts are made.