I am doing some practice problems to prepare for my statistics exam, and I just want to know if my reasoning is correct on one problem, and if not, I want to know how I should reason through this. The question is as follows:

X, Y are random variables following the Beta distribution B(120, 2019). What is P(2X+4 > 3Y)?

My reasoning is that since X,Y are following the Beta distribution, any x,y $\in$ X,Y must be bound to the interval (0,1). Thus, 2X+4 must be bound to (4,6), while 3Y must be bound to (0,3). So would I be right in saying this probability is 100% since min(X)>max(Y)? Thank you for any help. Again, I am not just looking for the answer, I am looking to see if my reasoning is correct, and how I should reason through it if my reasoning is incorrect.


Comment: Your logic seems correct; here is a quick simulation, using R statistical software.

set.seed(2019); m = 10^6
x = rbeta(m, 120, 2019); y = rbeta(m, 120, 2019)
mean(2*x + 4 > 3*y)
[1] 1   # TRUE for 100% of simulated values.

Also, you must have $D = 2X - 3Y > -4.$ Here is a summary of the simulated values of $D,$ all of which exceed $-4.$ Indeed the largest $D \approx 0.03, $ which agrees with what you say.

d = 2*x - 3*y
    Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
-0.15004 -0.06806 -0.05594 -0.05613 -0.04396  0.03038 
  • $\begingroup$ Another, slightly less attractive approach would be to notice that $\mathsf{Beta}(120, 2019)$ is approximately normal, then get a normal approximation of $D$ and show it has (essentially) all its probability above $-4.$ That might be an approach to approximate, say $P(-0.1 < D < -0.05) \approx 0.37,$ without using simulation. $\endgroup$ – BruceET Jan 21 '19 at 4:15

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.