HL Probability Notes
HL Probability Notes
Probability Notes
1 Prior Knowledge - Terms, symbols and notation
Probability theory is the study of the chance (or likelihood) of events happening.
Consider this: You throw a six-sided die and you want to find the probability of getting a 4.
You do this thirty times and record the number on the top face of the die each time.
Number of Trials (n) How many times you do the experiment; e.g. n = 30.
Outcome What happens each time; e.g. the number on the top face of the die.
Equally Likely All possible outcomes have the same chance of happening.
Sample Space (U) All the possible outcomes; e.g. U = {1, 2, 3, 4, 5, 6}.
n ( E) 1
Probability P( E)= ; e.g. P(4)= .
n(U ) 6
Experimental What you actually get when you do the experiment a number of times;
Probability e.g. If you threw the dice 30 times and 4 came up 7 times,
7
then P( 4)= .
30
NOTE: Experimental probability = Relative frequency
Frequency The number of times that an outcome is observed;
e.g. If 4 came up 7 times the frequency = 7.
Discrete Random Take on some value as the result of a counting process. The question,
Variables "How many?" is being asked. We can find the probability that a discrete
random variable is equal to a particular value.
e.g. X = The number of days with rain in a year.
2 Probability functions
Every random variable has an associated probability function, usually shown as a formula or
in a table.
For a discrete random variable, X say, we would have:
Example: Let X be the random variable representing the number of heads obtained when a
fair coin is tossed twice.
For this situation, U = { TT , TH , HT , HH } , and can now be represented in the following
table:
Values of x 0 1 2
P( X = x ) 0.25 0.5 0.25
3 Complementary events
Definition: Two events are called complementary if their union is the sample
space.
Given an event, A , and its complementary event, A' , we have the following:
• P( A) + P( A ' ) = 1 OR P( A) = 1 − P( A' ) .
P( R' ) = 1 − P(R)
= 1 − 0.35
= 0.65
4 Independent events
Definition: Two events A and B are independent if :
P( A ∩ B) = P( A) . P(B)
In practice, it means that the first event A should have no effect on the outcome of the
second event B .
Example: The probability that a car chosen at random is exceeding 50 km h-1 is 0.3.
The probability that a car chosen at random is without a current warrant of
fitness is 0.05. Find the probability that a car chosen at random is both
exceeding 50 km h-1 and does not have a current warrant of fitness.
Solution:
P(S ∩ W ' ) = P(S ) × P (W ' )
= 0.3 × 0.05
= 0.015
P( A ∪ B) = P( A) + P(B)
In practice, it means that the first event A should have nothing in common with the
second event B .
Example: The probability that a bank receives its daily computer printouts for a
previous day's transactions before 9:00 am is 0.12. The probability that it
receives it after 5:00 pm is 0.35. If office hours are from 9:00 am to 5:00 pm,
find the probability that, on a day chosen at random, the printout arrives
outside of office hours.
P( A ∪ B) = P( A) + P(B) − P( A ∩ B)
7 Probability Trees
• The tree branches from left to right.
• To find probabilities of events shown on the right, we multiply probabilities along the
branches leading to them.
• At each stage, probabilities must add vertically to 1.
Example 1: A basketball team plays 2 out of 5 games at home. If they play at home, the
probability of them winning is 0.7. If they are playing away, their chances of
winning are 0.5.
Find the probability that they win their game if it is not known whether the
game is going to be played at home or away.
Example 2: Roman and Jan play darts. They have alternate throws at the dart board.
The first person to hit 20 wins the game.
The probability of Roman hitting 20 on a single throw is 0.12.
The probability of Jan hitting 20 on a single throw is 0.16.
If Roman starts the game, find the probability that he wins.
Solution:
P( R wins) = P( R wins throw 1) + P( R wins throw 3) + P(R wins throw 5) + ... ...
= 0.12 + ( 0.88)(0.84)(0.12) + (0.88)(0.84)( 0.88)( 0.84)(0.12) + ... ...
u1
S∞ =
1−r
0.12
=
1 − (0.88)( 0.84)
= 0.460
Therefore, the probability that Roman wins the darts game is 0.460.
8 Conditional Probability
Some events cannot occur unless another event has occurred first. Such probabilities are
called conditional.
The probability of event A occurring is influenced by whether event B has occurred first.
The occurrence of event B may make event A certain, more likely, less likely or
impossible to happen.
The conditional probability that event A occurs, given that event B has occurred is
written P( A | B)
Multiplying along the branches of the probability tree gives us the following formula.
P( B) . P( A | B) = P( A ∩ B)
P(A ∩ B)
P(A | B) =
P(B)
Find the probability that if he plays a game of darts, he also plays a game of
snooker.
P(S ∩ D)
P( S | D) =
P(D)
0.20
=
0.65
= 0.308
Example 1: When a game of bridge is played (four players with 52 cards) and X is the
number of hearts in a players hand, we have:
13
E( X ) = Since there are 13 hearts in a normal pack of cards.
4
This is the theoretical average value of the number of
= 3.25 hearts in each players hand.
Example 2: A fair six-sided die is tossed once. Show that the expectation of X , the number
shown on the top face of the die, is 3.5 .
Solution: The probability distribution for a fair, six-sided die is given by:
Values of x 1 2 3 4 5 6
1 1 1 1 1 1
P( X = x )
6 6 6 6 6 6
1 1 1 1 1 1
E( X ) = 1( ) + 2( ) + 3( ) + 4( ) + 5( ) + 6( )
6 6 6 6 6 6
21
=
6
= 3.5
Note: We can deduce that:
E( a X ) = a E (X )
E( X + b) = E( X ) + b
E(a X + b) = a E ( X ) + b
Var(X) can be defined more formally as the expected squared distance from its mean; i.e.
it averages these distances out. It can be found using the following equivalent formulae:
Example 3: A single six-sided fair die is tossed once. Find the variance of the number of
dots shown on the top face of the die.
2 1 2 1 2 1 2 1 2 1 2 1 2
Var ( X ) = 1 ( ) + 2 ( ) + 3 ( ) + 4 ( ) + 5 ( ) + 6 ( ) − 3.5
6 6 6 6 6 6
= 2.92
Solutions:
( ) × ( 23 ) = 272
2
1
(a) P( GGR) =
3
() ()
2
1 2 1
= 3× × =
3 3 9
Extending this further, what would be the probability of getting 6 greens and 4 reds
from 10 spins?
() ()
6 4
1 2
Answer = ?? × × , where ?? is the number of ways of getting 6 greens
3 3
out of 10 spins.
() ()
6 4
1 2
Therefore, Answer = 10C6 × × .
3 3
Hence, the probability of getting r greens out of n spins will be given by
() ()
r n−r
n 1 2
Answer = Cr × × .
3 3
Therefore, the general formula is
P( X = x) = nC r p r q n − r
where X is the binomial random variable representing the number of successes,
n is the number of trials, p is the probability of success in an individual trial and
q is the probability of failure; that is, q = 1 − p .
Find the probability that three of the cars don't have a warrant of fitness.
1 5
Solution: We have n = 5 , p = and q = .
6 6
() ()
3 5−3
5 1 5
P( X = 3) = C3 × ×
6 6
10 × (
216 ) ( 36 )
1 25
= ×
125
=
3888
= 0.0322
Example 2: The organisers of the Singapore F1 Grand Prix know that 35% of racing cars
develop some sort of engine trouble during a race. If there are 15 cars in the
race, find the probability that at least 2 cars develop engine trouble during the
race.
P( X ≥ 2) 1 − P( X ≤ 1)
= 1 − ( P( X = 0) + P( X = 1))
= 1 − ( 15C0 × ( 0.35)0 × ( 0.65)15 + 15C1 × (0.35)1 × (0.65)14 )
= 1 − 0.0141787846
= 0.986
• Mean = E(X) = μ = n p .
• Variance = σ 2 = n p q .
• Standard Deviation = σ = √ n p q .
For a normal distribution with mean μ and standard deviation σ , the proportional
breakdown of where the random variable could lie is given below.
Example: The lengths of bolts made by a company are normally distributed with a
mean of 4 cm and a standard deviation of 0.05 cm.
(a) Find the probability that a bolt chosen at random will be between 3.9 cm and 4 cm.
(b) Find the probability that a bolt chosen at random will be less than 3.9 cm.
(c) Find the probability that a bolt chosen at random will be greater than 4.05 cm.
Solutions 1: (In a non-calculator paper, using what we know about areas and the number
of standard deviation from the mean.)
(a)
(b)
(c)
Note: You should always draw a quick sketch and shade the area you are trying to find.
A product has a normally distributed weight, with a mean 800 g and a standard
deviation of 100 g. Find the probability that the weight of a randomly chosen item
is between 800 g and 950 g.
x−μ .
z= σ
Therefore, to do the problem above, we need to use the standard normal distribution
and the z score.
z = x−μ
σ
950 − 800
=
100
= 1.5
Note: Using GDC where μ = 800 and σ = 100 with normalcdf(800, 100, 950), you will
also get 0.433 .
Note 1: The invNorm( function on your GDC uses the area from − ∞ to a in order to
find the value of a .
This is the white area in the diagram below.
Note 2: Sometimes we need to convert to z scores to find the answers. When trying to find
the value of μ or the value of σ , we always have to use z scores.
P ( Z > a ) = 0.8
Therefore, P ( Z ≤ a ) = 0.2 Using GDC with invNorm(0.2 , 0 , 1) which is the
white area in the diagram above.
Therefore, a = − 0.842 If we used invNorm(0.8 , 0 , 1), we would get + 0.842.
Example 2: A university professor determines that no more than 75% of this year's
History students should pass the final examination. The examination results
were approximately normally distributed with mean 62 and standard
deviation 13. Find the lowest score, k , necessary to pass the test.
We have μ = 62 and σ = 13 .
P ( X ≥ k ) = 0.75
Hence, P ( X < k ) = 0.25
Therefore, k = 53.23163... Using GDC with invNorm(0.25 , 62 , 13)
P ( X ≤ 58.2) = 0.15
Therefore,
(
P Z<
58.2 − μ
5.9 ) = 0.15 Converting to a z score using z =
x −μ
σ .
58.2 − μ
= − 1.03643... Using GDC with invNorm(0.15 , 0 , 1)
5.9
58.2 − μ = − 6.11495...
Therefore, μ = 64.3
A Poisson process is where discrete events occur within a continuous, but finite, interval of
time or space.
1. For a small interval, the probability of the event occurring is proportional to the size
of the interval.
2. The probability of more than one occurrence in the small interval is negligible.
3. Events must not occur simultaneously.
4. Each occurrence must be independent of others.
5. The event must occur at random.
Note: The Poisson random variable, X say, is the number of such occurrences within the
fixed interval. In theory, there is no upper limit on the number of such occurrences.
e− λ . λ x
f (x) =
x!
−5 3
P ( X = 3) = e .5
3!
= 0.140
Example 2: A 1200 m length of telephone cable has 10 faults in it. If a 100 m length of
similar cable is chosen at random:
(a) Find the mean and standard deviation of the number of faults per
100 m.
Solution: (a)
1
λ = × 10
12
5
=
6
σ =
√ 5
6
= 0.203
(b)
P ( X ≥ 2) = 1 − P( X ≤ 1)
= 1 − P( X = 0) −P ( X = 1)
() ()
5 0 5 1
−
6 5 −
6 5
e . e .
= 6 6
1− −
0! 1!
= 0.203
Example 3: Over a long period of time, the number of road accidents per day on a certain
road has been found to have approximately a Poisson Distribution with mean
2 on weekdays and 3 at the weekend.
(a) Find the probability that there are 3 accidents on a day chosen at
random.
(b) If there are no accidents on a day, find the probability that it is a
weekday.
a)
P ( X = 3) = .1289 + .0640
= .193 Rounding to 3 s.f.
b)
P( weekday ∩ no accidents )
P( weekday/no accidents) =
P(no accidents )
.0966
=
.0142 + .0966
= .872 Rounding to 3 s.f.
Example 4: In samples of material taken from a textile factory, 40% had at least one
fault. Estimate the mean number of faults per sample.
Solution: Let the random variable X be the number of faults per sample.
X has a Poisson distribution.
P( X ≥ 1) = 0.4
P( X = 0) = 0.6 Since P ( X = 0 ) = 1 − P( X ≥ 1)
−λ 0
e .λ
= 0.6
0!
e− λ = 0.6
ln(e− λ ) = ln(0.6)
− λ = ln(0.6)
λ = 0.512 Rounding to 3 s.f.