More general trees

It is straight forward to generalise the tree diagram in several ways without changing its properties (1-3). One possibility is to add more generations. Consider a random experiment with three events AA, BB, and CC. We can form the following tree:

Here,

p1=p(A) and p2=p(A)p_1=p(A) \text{ and }p_2=p(A^\prime)

and all other probabilities are conditional probabilities. E.g. p3p_3 is the probability that BB occurs, given that AA has occurred

p3=p(BA)p_3=p(B\vert A)

and p7p_7 is the probability that CC occurs given that AA and BB have occurred

p7=p(CAB)p_7=p(C|A \cap B)

The product of the branch probabilities p1p3p7p_1\cdot p_3\cdot p_7 is then the probability that AA, BB, and CC occurred simultaneously:

p(ABC)=p1p3p7p(A\cap B \cap C)=p_1\cdot p_3\cdot p_7

Another possibility to extend the tree is to use more than two branches per node. So, consider three events A1,A2A_1, A_2 and A3A_3 which form a partition of the sample space SS, and another three events B1,B2B_1, B_2 and B3B_3 which also form a partition of SS. Recall that a partition divides the sample space. Thus, whenever the experiment is performed, exactly one of the events A1,A2A_1, A_2 and A3A_3 will occur, and also exactly one of the events B1,B2B_1, B_2, and B3B_3 will occur. Thus

p(A1)+p(A2)+p(A3)=1p(A_1)+p(A_2)+p(A_3)=1

and

p(B1)+p(B2)+p(B3)=1p(B_1)+p(B_2)+p(B_3)=1

(as was the case for the tree with two branches per node, where the partitions were AA and AA^\prime, and also BB and BB^\prime). We draw the tree as follows:

And here are some exercises ...

Exercise 1
Q1

In a town, 30%30\% of the population reads newspaper A, of these 50%50\% read newspaper B, and of these 10%10\% read newspaper C. A person is selected at random from the town. What is the probability that the person reads all three newspapers?

Q2

A box contains 4 blue balls and 5 red balls. Four balls are selected at random without replacement. What is the probability for selecting

  1. exactly 44 blue balls.
  2. exactly 33 blue balls.
  3. no blue ball.
  4. a blue ball given that the previous three balls were red.
Q3

A box contains 4 blue balls and 5 red balls. Four balls are selected at random with replacement. What is the probability for selecting

  1. exactly 44 blue balls.
  2. exactly 33 blue balls.
  3. a blue ball given that the previous three balls were red.
Q4

A small brewery has three bottling machines. Machine A fills 40%40\% of all bottles, machine B and C fill 30%30\% each. 5%5\% of the bottles filled by A, 4%4\% of the bottles filled by B, and 3%3\% of the bottles filled by C are rejected for some reason. If a bottle is filled by A or B, what is the probability that it is rejected?

Q5

A box contains 2 blue balls, 3 red balls and 4 yellow balls. Two balls are selected at random without replacement. What is the probability that

  1. the selected balls have different colour?

  2. the first ball is yellow given that the last ball is blue?

Solution
A1
A2
A3
A4
A5