0
$\begingroup$

cenario: You have marbles of 2 colors, red and blue. You put one red and one blue marble in a bag and take one at random. The probability of getting a red marble is obviously 50%. But then you add 5 more blue marbles in the bag and take one at random again. The probability of getting red for that attempt alone is obviously 1/7, but how do I calculate the overall probability of getting a red marble at least once through both attempts whereas each attempt has a different probability?

Attempt 1: 2 marbles, 1 red, 1 blue

Attempt 2: 7 marbles, 2 red, 5 blue

Attempt 3: 10 marbles, 3 red, 7 blue

It doesn't take a mathematician to figure out the probabilities for each individual attempt, but in the given scenario, how is the overall probability calculated?

By the way, I already figured it out using pictures: 105/140 chance. But is there some formula I can use for such math problems like above?

$\endgroup$
3
  • $\begingroup$ Multiply the ratios, and then subtract from 1. $$\frac12\cdot\frac57\cdot\frac7{10}=\frac{35}{140}$$$$1-\frac{35}{140}=\frac{105}{140}$$Obviously, you need to reduce this fraction. $\endgroup$ Commented Nov 19, 2019 at 20:10
  • $\begingroup$ You can do this directly or you can do this indirectly depending on what is easiest at the time. For this, try finding the chance that you never drew a red marble. The probability you want is the opposite of this. Alternatively, doing it directly you can recall inclusion-exclusion tells us that $Pr(A\cup B\cup C) = Pr(A)+Pr(B)+Pr(C)-Pr(A\cap B)-Pr(A\cap C)-Pr(B\cap C)+Pr(A\cap B\cap C)$ $\endgroup$
    – JMoravitz
    Commented Nov 19, 2019 at 20:13
  • $\begingroup$ Alternatively still, you can approach with $Pr(A\cup B\cup C) = Pr(A)+Pr(A^c)Pr(B\mid A^c)+Pr(A^c)Pr(B^c\mid A^c)Pr(C\mid A^c\cap B^c)$. There are many different approaches. It is good to familiarize yourself with the most common ones as this lets you pick whatever is most suitable at the time. $\endgroup$
    – JMoravitz
    Commented Nov 19, 2019 at 20:14

2 Answers 2

1
$\begingroup$

Just note that $$1-p = (1-p_1)(1-p_2)(1-p_3)$$

This simply says that for the experiment to fail, you must fail on each attempt.

So the probability of success on the combined experiment is

$$p = 1-(1-p_1)(1-p_2)(1-p_3).$$

(Here, $p_j$ is the probability of success (drawing a red marble) on the $j^{\textrm{th}}$ attempt. I also assume you are drawing with replacement.)

$\endgroup$
0
$\begingroup$

Here you can break it into cases:

$$(\mbox{happens on first attempt})+(\mbox{doesn't happen on first attempt} )\cdot(\mbox{happens on second attempt} ) +(\mbox{doesn't happen on first or second attempt})\cdot (\mbox{happens on third attempt}) $$

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .