Three boxes with three coins (conditional probability)

568 Views Asked by At

So here is the problem,

There are 3 boxes and each box contains 3 coins. The first one has three golden coins, the second a golden one and two silvers and the third one has three silver coins. If we select a box randomly and it has a golden coin, what is the possibility that we will drag a golden coin from the same box again?

I tried to solve the problem like this:

$B_1=\{\text{the event of chosing the first box}\}$

$B_2=\{\text{the event of choosing the second box}\}$

$B_3=\{\text{the event of choosing the third box}\}$

$G=\{\text{the event of choosing a golden coin}\}$

$GG=\{\text{the event of choosing two golden coins in a row}\}$

We are actually looking for $$P(GG|G)=\frac{P(GG \cap G)}{P(G)}$$

So,

$P(G) = P(B_1)\times P(G_1)+P(B_2)\times P(G_2)+P(B_3)\times P(G_3)= \frac{1}{3}\times 1 + \frac{1}{3}\times \frac{1}{3} + \frac{1}{3}\times 0 = \frac{4}{9}$ (There are basically three different ways this can happen and I multiply the events leading to a golden coin in each way)

$P(GG) = P(B_1)\times P(G_1) \times P(GG_1)+P(B_2)\times P(G_2) \times P(GG_2)+P(B_3)\times P(G_3) \times P(GG_3)= \frac{1}{3}\times 1 \times 1 + \frac{1}{3}\times \frac{1}{3} \times 0 + \frac{1}{3}\times 0 \times 0= \frac{1}{3}$

(Again I multiply the events leading to the second golden coin considering that when we drag a golden coin, we don't replace it)

My difficulty here is calculating $P(GG \cap G)$

I tried $P(GG \cap G) = P(G)+P(GG) = \frac{4}{9} + \frac{1}{3} =\frac{7}{9}$

but $$P(GG|G)=\frac{\frac{7}{9}}{\frac{4}{9}}=\frac{7}{4}$$

Another attemp was $P(GG \cap G) = P(G|GG)P(GG) = \frac{1}{3} \times\frac{1}{3} =\frac{1}{9}$

and thus $$P(GG|G)=\frac{\frac{1}{9}}{\frac{4}{9}}=\frac{1}{4}$$ which intuitively feels correct but I don't understand the logic behind these two different approaches. Can someone please tell me if the second one is correct and why the first one isn't.

1

There are 1 best solutions below

0
On BEST ANSWER

The event of drawing two gold coins in a row includes the event of drawing the first cold coin. That is, you cannot draw two gold coins without drawing that first gold coin. Put differently, the $G$ event is not constraining the $GG$ in any way. Hence:

$$P(GG\cap G) = P(GG)$$