Specifically I want to find the probability that the product of two numbers chosen randomly on (0,1) interval is less than $ \frac{2}{9} $. The numbers are independent to each other.

That's what I done:

I define two random variables $X_1$ and $X_2$ which have Uniform distribution on $(0,1)$ interval.

Then the probability mass functions are

$f_{X_1}(x) = P(X_1 = x) = \left \{ \begin{matrix} 1 & \mbox{if x $\epsilon$ (0,1)} \\ 0 & \mbox{$otherwise$}\end{matrix}\right. $

$f_{X_2}(x) = P(X_2 = x) = \left \{ \begin{matrix} 1 & \mbox{if x $\epsilon$ (0,1)} \\ 0 & \mbox{$otherwise$}\end{matrix}\right. $

Solving the inequality equation

$ X_1X_2 \lt \frac{2}{9} \Rightarrow X_1 \lt \frac{2}{9X_2}$

Then integrating the mass probability functions

$P(X_1X_2 < \frac{2}{9}) = \displaystyle\int_{0}^{1} \displaystyle\int_{0}^{\frac{2}{9x_2}}\ f_{X_1}(x_1) f_{X_2}(x_2)dx_1dx_2 = \displaystyle\int_{0}^{1} \displaystyle\int_{0}^{\frac{2}{9x_2}}\ dx_1dx_2 = \frac{2}{9} \displaystyle\int_{0}^{1} \frac{dx_2}{x_2}$

Is easy to see that appears a Ln function. We can't solve Ln(0) so... What's wrong and how do I solve this? Thank you


When you got $X_1 \lt \frac 2{9X_2}$ you lost the fact that you also need $X_1 \lt 1$. You can't just use $\frac 2{9x_2}$ as the upper limit of the integral. In the range $0 \le x_2 \le \frac 29$ the upper limit is $1$. That removes the divergence.

  • $\begingroup$ Thanks! Good answer. Can you tell me if it is good? I need to do the sum of two double integrals. The first with $X_2$ in $(0, \frac{2}{9})$ and $X_1$ in $(0,1)$. The second with $X_2$ in $(\frac{2}{9}, 1)$ and $X_1$ in $(0, \frac{2}{9x_2})$ ? $\endgroup$ – DN_Euler Nov 27 '17 at 21:46
  • $\begingroup$ That is right. You shouldn't have any divergence problem any more. $\endgroup$ – Ross Millikan Nov 28 '17 at 1:21

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.