Entropic value at risk

From Wikipedia, the free encyclopedia

Jump to navigationJump to search

In financial mathematics and stochastic optimization, the concept of risk measure is used to
quantify the risk involved in a random outcome or risk position. Many risk measures have hitherto
been proposed, each having certain characteristics. The entropic value at risk (EVaR) is a coherent
risk measure introduced by Ahmadi-Javid,[1][2] which is an upper bound for the value at risk (VaR)
and the conditional value at risk (CVaR), obtained from the Chernoff inequality. The EVaR can also
be represented by using the concept of relative entropy. Because of its connection with the VaR
and the relative entropy, this risk measure is called "entropic value at risk". The EVaR was
developed to tackle some computational inefficiencies[clarification needed] of the CVaR. Getting
inspiration from the dual representation of the EVaR, Ahmadi-Javid[1][2] developed a wide class of
coherent risk measures, called g-entropic risk measures. Both the CVaR and the EVaR are
members of this class.

Contents

1 Definition

2 Properties

3 Examples

4 Optimization

5 Generalization (g-entropic risk measures)

6 See also

7 References

Definition

Let {\displaystyle (\Omega ,{\mathcal {F}},P)}(\Omega ,{\mathcal {F}},P) be a probability space with
{\displaystyle \Omega }\Omega a set of all simple events, {\displaystyle {\mathcal {F}}}{\mathcal
{F}} a {\displaystyle \sigma }\sigma -algebra of subsets of {\displaystyle \Omega }\Omega and
{\displaystyle P}P a probability measure on {\displaystyle {\mathcal {F}}}{\mathcal {F}}. Let
{\displaystyle X}X be a random variable and {\displaystyle \mathbf {L} _{M^{+}}}{\displaystyle
\mathbf {L} _{M^{+}}} be the set of all Borel measurable functions {\displaystyle X:\Omega \to
\mathbb {R} }{\displaystyle X:\Omega \to \mathbb {R} } whose moment-generating function
{\displaystyle M_{X}(z)}{\displaystyle M_{X}(z)} exists for all {\displaystyle z\geq 0}{\displaystyle
z\geq 0}. The entropic value at risk (EVaR) of {\displaystyle X\in \mathbf {L} _{M^{+}}}{\displaystyle
X\in \mathbf {L} _{M^{+}}} with confidence level {\displaystyle 1-\alpha }1-\alpha is defined as
follows:
{\displaystyle {\text{EVaR}}_{1-\alpha }(X):=\inf _{z>0}\left\{z^{-1}\ln \left({\frac {M_{X}(z)}
{\alpha }}\right)\right\}.}{\displaystyle {\text{EVaR}}_{1-\alpha }(X):=\inf _{z>0}\left\{z^{-1}\ln
\left({\frac {M_{X}(z)}{\alpha }}\right)\right\}.}

(1)

In finance, the random variable {\displaystyle X\in \mathbf {L} _{M^{+}},}{\displaystyle X\in
\mathbf {L} _{M^{+}},} in the above equation, is used to model the losses of a portfolio.

Consider the Chernoff inequality

{\displaystyle \Pr(X\geq a)\leq e^{-za}M_{X}(z),\quad \forall z>0.}{\displaystyle \Pr(X\geq a)\leq

e^{-za}M_{X}(z),\quad \forall z>0.}

(2)
Solving the equation {\displaystyle e^{-za}M_{X}(z)=\alpha }{\displaystyle e^{-za}M_{X}(z)=\alpha }
for {\displaystyle a,}a, results in

{\displaystyle a_{X}(\alpha ,z):=z^{-1}\ln \left({\frac {M_{X}(z)}{\alpha }}\right).}{\displaystyle a_{X}

(\alpha ,z):=z^{-1}\ln \left({\frac {M_{X}(z)}{\alpha }}\right).}

By considering the equation (1), we see that

{\displaystyle {\text{EVaR}}_{1-\alpha }(X):=\inf _{z>0}\{a_{X}(\alpha ,z)\},}{\displaystyle

{\text{EVaR}}_{1-\alpha }(X):=\inf _{z>0}\{a_{X}(\alpha ,z)\},}

which shows the relationship between the EVaR and the Chernoff inequality. It is worth noting
that {\displaystyle a_{X}(1,z)}{\displaystyle a_{X}(1,z)} is the entropic risk measure or exponential
premium, which is a concept used in finance and insurance, respectively.

Let {\displaystyle \mathbf {L} _{M}}{\displaystyle \mathbf {L} _{M}} be the set of all Borel
measurable functions {\displaystyle X:\Omega \to \mathbb {R} }{\displaystyle X:\Omega \to
\mathbb {R} } whose moment-generating function {\displaystyle M_{X}(z)}{\displaystyle M_{X}(z)}
exists for all {\displaystyle z}z. The dual representation (or robust representation) of the EVaR is as
follows:

{\displaystyle {\text{EVaR}}_{1-\alpha }(X)=\sup _{Q\in \Im }(E_{Q}(X)),}{\displaystyle

{\text{EVaR}}_{1-\alpha }(X)=\sup _{Q\in \Im }(E_{Q}(X)),}

(3)
where {\displaystyle X\in \mathbf {L} _{M},}{\displaystyle X\in \mathbf {L} _{M},} and {\displaystyle
\Im }\Im is a set of probability measures on {\displaystyle (\Omega ,{\mathcal {F}})}
(\Omega,\mathcal{F}) with {\displaystyle \Im =\{Q\ll P:D_{KL}(Q||P)\leq -\ln \alpha \}}
{\displaystyle \Im =\{Q\ll P:D_{KL}(Q||P)\leq -\ln \alpha \}}. Note that

{\displaystyle D_{KL}(Q||P):=\int {\frac {dQ}{dP}}\left(\ln {\frac {dQ}{dP}}\right)dP}{\displaystyle

D_{KL}(Q||P):=\int {\frac {dQ}{dP}}\left(\ln {\frac {dQ}{dP}}\right)dP}

is the relative entropy of {\displaystyle Q}Q with respect to {\displaystyle P,}P, also called the
Kullback–Leibler divergence. The dual representation of the EVaR discloses the reason behind its
naming.

Properties

The EVaR is a coherent risk measure.

The moment-generating function {\displaystyle M_{X}(z)}{\displaystyle M_{X}(z)} can be

represented by the EVaR: for all {\displaystyle X\in \mathbf {L} _{M^{+}}}{\displaystyle X\in
\mathbf {L} _{M^{+}}} and {\displaystyle z>0}z>0

{\displaystyle M_{X}(z)=\sup _{0<\alpha \leq 1}\{\alpha \exp(z{\text{EVaR}}_{1-\alpha }(X))\}.}

{\displaystyle M_{X}(z)=\sup _{0<\alpha \leq 1}\{\alpha \exp(z{\text{EVaR}}_{1-\alpha }(X))\}.}

(4)

For {\displaystyle X,Y\in \mathbf {L} _{M}}{\displaystyle X,Y\in \mathbf {L} _{M}}, {\displaystyle
{\text{EVaR}}_{1-\alpha }(X)={\text{EVaR}}_{1-\alpha }(Y)}{\displaystyle {\text{EVaR}}_{1-\alpha }
(X)={\text{EVaR}}_{1-\alpha }(Y)} for all {\displaystyle \alpha \in ]0,1]}{\displaystyle \alpha \in ]0,1]}
if and only if {\displaystyle F_{X}(b)=F_{Y}(b)}{\displaystyle F_{X}(b)=F_{Y}(b)} for all {\displaystyle
b\in \mathbb {R} }{\displaystyle b\in \mathbb {R} }.
The entropic risk measure with parameter {\displaystyle \theta ,}\theta , can be represented by
means of the EVaR: for all {\displaystyle X\in \mathbf {L} _{M^{+}}}{\displaystyle X\in \mathbf {L}
_{M^{+}}} and {\displaystyle \theta >0}\theta >0

{\displaystyle \theta ^{-1}\ln M_{X}(\theta )=a_{X}(1,\theta )=\sup _{0<\alpha \leq 1}\

{{\text{EVaR}}_{1-\alpha }(X)+\theta ^{-1}\ln \alpha \}.}{\displaystyle \theta ^{-1}\ln M_{X}(\theta )
=a_{X}(1,\theta )=\sup _{0<\alpha \leq 1}\{{\text{EVaR}}_{1-\alpha }(X)+\theta ^{-1}\ln \alpha \}.}

(5)

The EVaR with confidence level {\displaystyle 1-\alpha }1-\alpha is the tightest possible upper
bound that can be obtained from the Chernoff inequality for the VaR and the CVaR with
confidence level {\displaystyle 1-\alpha }1-\alpha ;

{\displaystyle {\text{VaR}}(X)\leq {\text{CVaR}}(X)\leq {\text{EVaR}}(X).}{\displaystyle {\text{VaR}}

(X)\leq {\text{CVaR}}(X)\leq {\text{EVaR}}(X).}

(6)
The following inequality holds for the EVaR:

{\displaystyle {\text{E}}(X)\leq {\text{EVaR}}_{1-\alpha }(X)\leq {\text{esssup}}(X)}{\displaystyle

{\text{E}}(X)\leq {\text{EVaR}}_{1-\alpha }(X)\leq {\text{esssup}}(X)}

(7)

where {\displaystyle {\text{E}}(X)}{\displaystyle {\text{E}}(X)} is the expected value of {\displaystyle

X}X and {\displaystyle {\text{esssup}}(X)}{\displaystyle {\text{esssup}}(X)} is the essential
supremum of {\displaystyle X}X, i.e., {\displaystyle \inf _{t\in \mathbb {R} }\{t:\Pr(X\leq t)=1\}}
{\displaystyle \inf _{t\in \mathbb {R} }\{t:\Pr(X\leq t)=1\}}. So do hold {\displaystyle
{\text{EVaR}}_{0}(X)={\text{E}}(X)}{\displaystyle {\text{EVaR}}_{0}(X)={\text{E}}(X)} and
{\displaystyle \lim _{\alpha \to 0}{\text{EVaR}}_{1-\alpha }(X)={\text{esssup}}(X)}{\displaystyle \lim
_{\alpha \to 0}{\text{EVaR}}_{1-\alpha }(X)={\text{esssup}}(X)}.

Examples

Comparing the VaR, CVaR and EVaR for the standard normal distribution

Comparing the VaR, CVaR and EVaR for the uniform distribution over the interval (0,1)

For {\displaystyle X\sim N(\mu ,\sigma ^{2}),}{\displaystyle X\sim N(\mu ,\sigma ^{2}),}

{\displaystyle {\text{EVaR}}_{1-\alpha }(X)=\mu +\sigma {\sqrt {-2\ln \alpha }}.}{\displaystyle

{\text{EVaR}}_{1-\alpha }(X)=\mu +\sigma {\sqrt {-2\ln \alpha }}.}
(8)

For {\displaystyle X\sim U(a,b),}{\displaystyle X\sim U(a,b),}

{\displaystyle {\text{EVaR}}_{1-\alpha }(X)=\inf _{t>0}\left\lbrace t\ln \left(t{\frac {e^{t^{-1}b}-

e^{t^{-1}a}}{b-a}}\right)-t\ln \alpha \right\rbrace .}{\displaystyle {\text{EVaR}}_{1-\alpha }(X)=\inf
_{t>0}\left\lbrace t\ln \left(t{\frac {e^{t^{-1}b}-e^{t^{-1}a}}{b-a}}\right)-t\ln \alpha \right\rbrace .}

(9)

Figures 1 and 2 show the comparing of the VaR, CVaR and EVaR for {\displaystyle N(0,1)}N(0,1)
and {\displaystyle U(0,1)}U(0,1).

Optimization

Let {\displaystyle \rho }\rho be a risk measure. Consider the optimization problem

{\displaystyle \min _{{\boldsymbol {w}}\in {\boldsymbol {W}}}\rho (G({\boldsymbol {w}},

{\boldsymbol {\psi }})),}{\displaystyle \min _{{\boldsymbol {w}}\in {\boldsymbol {W}}}\rho
(G({\boldsymbol {w}},{\boldsymbol {\psi }})),}
(10)

where {\displaystyle {\boldsymbol {w}}\in {\boldsymbol {W}}\subseteq \mathbb {R} ^{n}}

{\displaystyle {\boldsymbol {w}}\in {\boldsymbol {W}}\subseteq \mathbb {R} ^{n}} is an
{\displaystyle n}n-dimensional real decision vector, {\displaystyle {\boldsymbol {\psi }}}
{\displaystyle {\boldsymbol {\psi }}} is an {\displaystyle m}m-dimensional real random vector with
a known probability distribution and the function {\displaystyle G({\boldsymbol {w}},.):\mathbb
{R} ^{m}\to \mathbb {R} }{\displaystyle G({\boldsymbol {w}},.):\mathbb {R} ^{m}\to \mathbb {R} }
is a Borel measurable function for all values {\displaystyle {\boldsymbol {w}}\in {\boldsymbol
{W}}.}{\displaystyle {\boldsymbol {w}}\in {\boldsymbol {W}}.} If {\displaystyle \rho ={\text{EVaR}},}
{\displaystyle \rho ={\text{EVaR}},} then the optimization problem (10) turns into:

{\displaystyle \min _{{\boldsymbol {w}}\in {\boldsymbol {W}},t>0}\left\{t\ln M_{G({\boldsymbol

{w}},{\boldsymbol {\psi }})}(t^{-1})-t\ln \alpha \right\}.}{\displaystyle \min _{{\boldsymbol {w}}\in
{\boldsymbol {W}},t>0}\left\{t\ln M_{G({\boldsymbol {w}},{\boldsymbol {\psi }})}(t^{-1})-t\ln
\alpha \right\}.}

(11)
Let {\displaystyle {\boldsymbol {S}}_{\boldsymbol {\psi }}}{\displaystyle {\boldsymbol
{S}}_{\boldsymbol {\psi }}} be the support of the random vector {\displaystyle {\boldsymbol
{\psi }}.}{\displaystyle {\boldsymbol {\psi }}.} If {\displaystyle G(.,{\boldsymbol {s}})}{\displaystyle
G(.,{\boldsymbol {s}})} is convex for all {\displaystyle {\boldsymbol {s}}\in {\boldsymbol
{S}}_{\boldsymbol {\psi }}}{\displaystyle {\boldsymbol {s}}\in {\boldsymbol {S}}_{\boldsymbol
{\psi }}}, then the objective function of the problem (11) is also convex. If {\displaystyle
G({\boldsymbol {w}},{\boldsymbol {\psi }})}{\displaystyle G({\boldsymbol {w}},{\boldsymbol
{\psi }})} has the form

{\displaystyle G({\boldsymbol {w}},{\boldsymbol {\psi }})=g_{0}({\boldsymbol {w}})+\sum

_{i=1}^{m}g_{i}({\boldsymbol {w}})\psi _{i},\qquad g_{i}:\mathbb {R} ^{n}\to \mathbb {R}
,i=0,1,\ldots ,m,}{\displaystyle G({\boldsymbol {w}},{\boldsymbol {\psi }})=g_{0}({\boldsymbol {w}})
+\sum _{i=1}^{m}g_{i}({\boldsymbol {w}})\psi _{i},\qquad g_{i}:\mathbb {R} ^{n}\to \mathbb {R}
,i=0,1,\ldots ,m,}

(12)

and {\displaystyle \psi _{1},\ldots ,\psi _{m}}{\displaystyle \psi _{1},\ldots ,\psi _{m}} are
independent random variables in {\displaystyle \mathbf {L} _{M}}{\displaystyle \mathbf {L} _{M}},
then (11) becomes

{\displaystyle \min _{{\boldsymbol {w}}\in {\boldsymbol {W}},t>0}\left\lbrace g_{0}({\boldsymbol

{w}})+t\sum _{i=1}^{m}\ln M_{g_{i}({\boldsymbol {w}})\psi _{i}}(t^{-1})-t\ln \alpha \right\rbrace .}
{\displaystyle \min _{{\boldsymbol {w}}\in {\boldsymbol {W}},t>0}\left\lbrace g_{0}({\boldsymbol
{w}})+t\sum _{i=1}^{m}\ln M_{g_{i}({\boldsymbol {w}})\psi _{i}}(t^{-1})-t\ln \alpha \right\rbrace .}
(13)

which is computationally tractable. But for this case, if one uses the CVaR in problem (10), then
the resulting problem becomes as follows:

{\displaystyle \min _{{\boldsymbol {w}}\in {\boldsymbol {W}},t\in \mathbb {R} }\left\lbrace t+

{\frac {1}{\alpha }}{\text{E}}\left[g_{0}({\boldsymbol {w}})+\sum _{i=1}^{m}g_{i}({\boldsymbol
{w}})\psi _{i}-t\right]_{+}\right\rbrace .}{\displaystyle \min _{{\boldsymbol {w}}\in {\boldsymbol
{W}},t\in \mathbb {R} }\left\lbrace t+{\frac {1}{\alpha }}{\text{E}}\left[g_{0}({\boldsymbol {w}})
+\sum _{i=1}^{m}g_{i}({\boldsymbol {w}})\psi _{i}-t\right]_{+}\right\rbrace .}

(14)

It can be shown that by increasing the dimension of {\displaystyle \psi }\psi , problem (14) is
computationally intractable even for simple cases. For example, assume that {\displaystyle \psi
_{1},\ldots ,\psi _{m}}{\displaystyle \psi _{1},\ldots ,\psi _{m}} are independent discrete random
variables that take {\displaystyle k}k distinct values. For fixed values of {\displaystyle {\boldsymbol
{w}}}{\boldsymbol {w}} and {\displaystyle t,}t, the complexity of computing the objective function
given in problem (13) is of order {\displaystyle mk}mk while the computing time for the objective
function of problem (14) is of order {\displaystyle k^{m}}k^{m}. For illustration, assume that
{\displaystyle k=2,m=100}{\displaystyle k=2,m=100} and the summation of two numbers takes
{\displaystyle 10^{-12}}10^{-12} seconds. For computing the objective function of problem (14)
one needs about {\displaystyle 4\times 10^{10}}4\times 10^{10} years, whereas the evaluation of
objective function of problem (13) takes about {\displaystyle 10^{-10}}10^{{-10}} seconds. This
shows that formulation with the EVaR outperforms the formulation with the CVaR (see [2] for
more details).

Generalization (g-entropic risk measures)

Drawing inspiration from the dual representation of the EVaR given in (3), one can define a wide
class of information-theoretic coherent risk measures, which are introduced in.[1][2] Let
{\displaystyle g}g be a convex proper function with {\displaystyle g(1)=0}{\displaystyle g(1)=0} and
{\displaystyle \beta }\beta be a non-negative number. The {\displaystyle g}g-entropic risk measure
with divergence level {\displaystyle \beta }\beta is defined as

{\displaystyle {\text{ER}}_{g,\beta }(X):=\sup _{Q\in \Im }{\text{E}}_{Q}(X)}{\displaystyle

{\text{ER}}_{g,\beta }(X):=\sup _{Q\in \Im }{\text{E}}_{Q}(X)}

(15)

where {\displaystyle \Im =\{Q\ll P:H_{g}(P,Q)\leq \beta \}}{\displaystyle \Im =\{Q\ll P:H_{g}
(P,Q)\leq \beta \}} in which {\displaystyle H_{g}(P,Q)}{\displaystyle H_{g}(P,Q)} is the generalized
relative entropy of {\displaystyle Q}Q with respect to {\displaystyle P}P. A primal representation of
the class of {\displaystyle g}g-entropic risk measures can be obtained as follows:
{\displaystyle {\text{ER}}_{g,\beta }(X)=\inf _{t>0,\mu \in \mathbb {R} }\left\lbrace t\left[\mu +
{\text{E}}_{P}\left(g^{*}\left({\frac {X}{t}}-\mu +\beta \right)\right)\right]\right\rbrace }
{\displaystyle {\text{ER}}_{g,\beta }(X)=\inf _{t>0,\mu \in \mathbb {R} }\left\lbrace t\left[\mu +
{\text{E}}_{P}\left(g^{*}\left({\frac {X}{t}}-\mu +\beta \right)\right)\right]\right\rbrace }

(16)

where {\displaystyle g^{*}}g^{*} is the conjugate of {\displaystyle g}g. By considering

{\displaystyle g(x)={\begin{cases}x\ln x&x>0\\0&x=0\\+\infty &x<0\end{cases}}}{\displaystyle

g(x)={\begin{cases}x\ln x&x>0\\0&x=0\\+\infty &x<0\end{cases}}}

(17)

with {\displaystyle g^{*}(x)=e^{x-1}}g^*(x)=e^{x-1} and {\displaystyle \beta =-\ln \alpha }

{\displaystyle \beta =-\ln \alpha }, the EVaR formula can be deduced. The CVaR is also a
{\displaystyle g}g-entropic risk measure, which can be obtained from (16) by setting
{\displaystyle g(x)={\begin{cases}0&0\leq x\leq {\frac {1}{\alpha }}\\+\infty
&{\text{otherwise}}\end{cases}}}{\displaystyle g(x)={\begin{cases}0&0\leq x\leq {\frac {1}
{\alpha }}\\+\infty &{\text{otherwise}}\end{cases}}}

(18)

with {\displaystyle g^{*}(x)={\tfrac {1}{\alpha }}\max\{0,x\}}{\displaystyle g^{*}(x)={\tfrac {1}

{\alpha }}\max\{0,x\}} and {\displaystyle \beta =0}\beta =0 (see [1][3] for more details).

For more results on {\displaystyle g}g-entropic risk measures see.[4]

See also

Stochastic optimization

Risk measure

Coherent risk measure

Value at risk

Conditional value at risk

Expected shortfall

Entropic risk measure

Kullback–Leibler divergence

Generalized relative entropy

References
Ahmadi-Javid, Amir (2011). An information-theoretic approach to constructing coherent risk
measures. St. Petersburg, Russia: Proceedings of IEEE International Symposium on Information
Theory. pp. 2125–2127. doi:10.1109/ISIT.2011.6033932.

Ahmadi-Javid, Amir (2012). "Entropic value-at-risk: A new coherent risk measure". Journal of
Optimization Theory and Applications. 155 (3): 1105–1123. doi:10.1007/s10957-011-9968-2.

Ahmadi-Javid, Amir (2012). "Addendum to: Entropic Value-at-Risk: A New Coherent Risk
Measure". Journal of Optimization Theory and Applications. 155 (3): 1124–1128.
doi:10.1007/s10957-012-0014-9.

Breuer, Thomas; Csiszar, Imre (2013). "Measuring Distribution Model Risk". arXiv:1301.4832v1.