Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Suppose $x$ and $y$ are jointly normal with known parameters. I need to find the probability of the event in the parenthesis: $\Pr\left(x>f(y; a),y>g(x;b)\right)$. The function $f(\cdot)$ is increasing, differentiable and invertible and $a$ is a parameter. $g(\cdot)$ is decreasing, differentiable and invertible while $b$ is a parameter. Thank you

share|cite|improve this question
up vote 1 down vote accepted

Since $f(\cdot)$ is increasing and invertible $X > f(Y;a)$ implies $f^{(-1)}(X;a) > Y$, thus: $$ \begin{eqnarray} \Pr\left(X>f(Y; a),Y>g(X;b)\right) &=& \Pr\left(f^{(-1)}(X;a)>Y>g(X;b)\right) \\ &=& \mathbb{E}\left(\Pr\left(f^{(-1)}(X;a)>Y>g(X;b) \mid X \right)\right) \\ &=& \mathbb{E}\left( \Phi_{Y|X}\left(f^{(-1)}(X;a)\right) - \Phi_{Y|X} \left( g(X;b) \right) \right) \end{eqnarray} $$ Now for $(X,Y)$ jointly normal with $\mathbb{E}(X) = m_1$, $\mathbb{E}(Y)=m_2$, $\mathbb{Var}(X) = \sigma_1^2$, $\mathbb{Var}(Y) = \sigma_2^2$, $\mathbb{Cor}\left(X, Y\right)=\rho$, $$ Y \mid X \sim \mathcal{N}\left(m_2 + \frac{\sigma_2}{\sigma_1} \rho \left(X-m_1\right), \sigma_2 \sqrt{1-\rho^2}\right) $$ Thus $$\Phi_{Y|X}\left(Z\right) = \Phi\left(\frac{1}{ \sqrt{1-\rho}} \left( \frac{1}{\sigma_2} \left(Z-m_2\right) + \frac{\rho}{\sigma_1} \left(X-m_1\right) \right) \right) $$ Now substitute back and cross your fingers that $f^{(-1)}$ and $g$ are simple enough to permit analytic evaluation. Otherwise numeric methods can be used.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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