Risk measure

Not to be confused with deviation risk measures, e.g. standard deviation .

A Risk measure is used to determine the amount of an asset or set of assets (traditionally currency) to be kept in reserve. The purpose of this reserve is to make the risks taken by financial institutions, such as banks and insurance companies, acceptable to the regulator. In recent years attention has turned towards convex and coherent risk measurement.

Mathematically

A risk measure is defined as a mapping from a set of random variables to the real numbers. This set of random variables represents the risk at hand. The common notation for a risk measure associated with a random variable $X$ is $\rho[X]$. A risk measure $\rho: \mathcal{L} \to \mathbb{R} \cup \{+\infty\}$ should have certain properties:

Normalized

$\rho(0) = 0$

Translative

$\mathrm{If}\; a \in \mathbb{R} \; \mathrm{and} \; Z \in \mathcal{L} ,\;\mathrm{then}\; \rho(Z + a) = \rho(Z) - a$

Monotone

$\mathrm{If}\; Z_1,Z_2 \in \mathcal{L} \;\mathrm{and}\; Z_1 \leq Z_2 ,\; \mathrm{then} \; \rho(Z_1) \geq \rho(Z_2)$

Set-valued

In a situation with $\mathbb{R}^d$-valued portfolios such that risk can be measured in $m \leq d$ of the assets, then a set of portfolios is the proper way to depict risk. Set-valued risk measures are useful for markets with transaction costs.

Mathematically

A set-valued risk measure is a function $R: L_d^p \rightarrow \mathbb{F}_M$, where $L_d^p$ is a $d$-dimensional Lp space, $\mathbb{F}_M = \{D \subseteq M: D = cl (D + K_M)\}$, and $K_M = K \cap M$ where $K$ is a constant solvency cone and $M$ is the set of portfolios of the $m$ reference assets. $R$ must have the following properties:

Normalized

$K_M \subseteq R(0) \; \mathrm{and} \; R(0) \cap -\mathrm{int}K_M = \emptyset$

Translative in M

$\forall X \in L_d^p, \forall u \in M: R(X + u1) = R(X) - u$

Monotone

$\forall X_2 - X_1 \in L_d^p(K) \Rightarrow R(X_2) \supseteq R(X_1)$

Examples

Well known risk measures

• Value at risk
• Expected shortfall
• Tail conditional expectation
• Entropic risk measure
• Superhedging price
• ...

Variance

Variance (or standard deviation) is not a risk measure. This can be seen since it has neither the translation property or monotonicity. That is $Var(X + a) = Var(X) \neq Var(X) - a$ for all $a \in \mathbb{R}$, and a simple counterexample for monotonicity can be found. The standard deviation is a deviation risk measure.

Relation to Acceptance Set

There is a one-to-one correspondence between an acceptance set and a corresponding risk measure. As defined below it can be shown that $R_{A_R}(X) = R(X)$ and $A_{R_A} = A$.

Risk Measure to Acceptance Set

• If $\rho$ is a (scalar) risk measure then $A_{\rho} = \{X \in L^p: \rho(X) \leq 0\}$ is an acceptance set.
• If $R$ is a set-valued risk measure then $A_R = \{X \in L^p_d: 0 \in R(X)\}$ is an acceptance set.

Acceptance Set to Risk Measure

• If $A$ is an acceptance set (in 1-d) then $\rho_A(X) = \inf\{u \in \mathbb{R}: X + u1 \in A\}$ defines a (scalar) risk measure.
• If $A$ is an acceptance set then $R_A(X) = \{u \in M: X + u1 \in A\}$ is a set-valued risk measure.

Relation with deviation risk measure

There is a one-to-one relationship between a deviation risk measure D and an expectation-bounded risk measure $\rho$ where for any $X \in \mathcal{L}^2$

• $D(X) = \rho(X - \mathbb{E}[X])$
• $\rho(X) = D(X) - \mathbb{E}[X]$.

$\rho$ is expectation bounded if $\rho(X) > \mathbb{E}[-X]$ for any nonconstant X and $\rho(X) = \mathbb{E}[-X]$ for any constant X.

See also

• Dynamic risk measure
• Managerial risk accounting
• Risk management
• RiskMetrics
• Spectral risk measure

Further reading

• Crouhy, Michel; D. Galai, and R. Mark (2001). Risk Management. McGraw-Hill. pp. 752 pages. ISBN 0-07-135731-9. 
• Kevin, Dowd (2005). Measuring Market Risk (2nd ed.). John Wiley & Sons. pp. 410 pages. ISBN 0-470-01303-6.