0
$\begingroup$

Here is the definition of VaR (Value at Risk) taken from McNeil, Alexander J., Rüdiger Frey and Paul Embrechts (2015), Quantitative risk management: Concepts, techniques and tools: $$ \textrm{VaR}_{\alpha}(L)=\inf\{l\in\mathbb{R}\colon P(L>l)\leq 1-\alpha\} =\inf\{l\in\mathbb{R}\colon F_{L}(l)\geq\alpha\} $$ where $L$ is the Loss distribution and $F_L$ its cumulative distribution function.

I would like to translate it in terms of Profit & Loss distribution $X=-L$ but I'm a bit unsure this is correct: $$\textrm{VaR}_{\alpha}(X)=-\inf\{x\in\mathbb{R}\colon P(X<x)\leq 1-\alpha\}$$ $$ = -\inf\{x\in\mathbb{R}\colon F_{X}(x)\geq 1-\alpha\}$$

Obviously, the two values must be the same, that is:

$$\textrm{VaR}_{\alpha}(L)=\textrm{VaR}_{\alpha}(X)=\textrm{VaR}_{\alpha}$$

Could you please check my formula? In particular, should I use $\sup$ instead of $\inf$?

$\endgroup$
0
$\begingroup$

I drew some graphs, and now I think I can answer my own question.

$$\textrm{VaR}_{\alpha}(X)=-\inf\{x\in\mathbb{R}\colon P(X<x)\leq 1-\alpha\}$$

is wrong because I have to take the upper bound, since the set is made of negative numbers, so $$\textrm{VaR}_{\alpha}(X)=-\sup\{x\in\mathbb{R}\colon P(X<x)\leq 1-\alpha\}$$ is the correct formula.

$$\textrm{VaR}_{\alpha}(X)= -\inf\{x\in\mathbb{R}\colon F_{X}(x)\geq 1-\alpha\}$$ is correct, but to make it equivalent to the formula with the Loss distribution, even when there is no minimum, it should be: $$\textrm{VaR}_{\alpha}(X)= -\sup\{x\in\mathbb{R}\colon F_{X}(x)\leq 1-\alpha\}$$

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.