UNIT 4: RETURN OF THE LECTURES
PROBABILITY IS HERE
MATHS METHODS
FT. BRACKETS
& BRACKETS
& BRACKETS
NO
SPAGHETTI
HERE
LECTURE OUTLINE
Integration
Discrete Probability
Binomial Distributions
Normal Distributions
Continuous Probability
Statistics
LEARNING OUTCOMES
Antidifferentiation and Integration
Computations
Discrete Probability
Probability properties
Sample spaces
Conditional probability, independence and mutual
exclusiveness
Discrete random variables
Expectancy and Variance
INTEGRATION
Integral, heh, to our learning of continuous probability.
Is the opposite function to the derivative
= +
Dont forget and dont forget +
BASIC RULES
RULE EXAMPLE
1
= +1 + 2 = 2 +
+1
1 +1 2
1 3
+ = + + 2 + 3 = 2 + 3 +
+1 6
1
= + 2 2 = 2 +
1 3
= log + = 3 log +
1
sin = cos + 2 sin = 2 cos +
1 1
cos = sin + cos 3 = sin 3 +
3
INTEGRATION
Integrating between two bounds is called integration
We apply the fundamental theorem of calculus
= ()
Where () is an antiderivative of ()
We can drop the + but not the
INTEGRATION EXAMPLE
2
sin 4 + 4 3
0
1 2
= cos 4 + 4
4 0
1 4 4 1
= cos + cos 0 + 0 4
4 2 2 4
1 4 1
= + +
4 16 4
4
=
16
PROBABILITY
Probability is the extent to which an event is likely to occur
Is measured bye the ratio of favourable outcomes to total
outcomes
number of
Pr =
total
We measure probability in decimals such that
0 Pr() 1
SAMPLE SPACE
A sample space is a list of the total possible outcomes
Consider two fair six sided dice, the sample space is as
follows:
1,1 , 1,2 , 1,3 6,6
Consider a trio of fair coins being tossed, the sample space
is as follows:
, , , , , , , , , , ,
PROBABILITY PROPERTIES
We represent common entities between sample spaces as
intersections
Pr
We represent all of the entities between sample spaces as
the union
Pr = Pr + Pr Pr( )
VENN DIAGRAMS
Pr(
)
KARNAUGH MAPS
Pr = 1 Pr
Read as: A complement, meaning all but
Pr( ) Pr( ) Pr()
Pr( ) Pr( ) Pr( )
Pr() Pr( ) 1
KARNAUGH MAPS
In a sample space:
Pr = 0.1 and Pr = 0.2 and Pr A = 0.1
0.1 0.1 0.2
0 0.8 0.8
0.1 0.9 1
TREE DIAGRAMS
Tree diagrams are useful for multi-step probability
problems, that is, problems that require multiple stages,
such as flipping a coin and then flipping another one.
Pr = 0.5 0.3 = 0.15
0.3
0.5
0.7
Pr = 0.5 0.7 = 0.35
0.5
PROBABILITY PROPERTIES
The probability of an event occurring, given that a certain
event has already happened is called conditional
Pr( )
Pr =
Pr()
If two events are mutually exclusive, then they have no
intersection, thus their union is their added probabilities
Pr = Pr + Pr()
PROBABILITY PROPERTIES
If two events are independent, that is one event occurring
has no impact on the other, they have the following
properties
Pr = Pr()
Pr = Pr Pr()
Mutual exclusivity should not be confused with
independence, they have different properties and one
existing does not necessarily mean the other does as well
DISCRETE RANDOM VARIABLES
A discrete random variable has the following properties:
It can only take countable values
The probabilities add to one
Which of the following are discrete variables?
Money Gender Postcodes
Months in progress Number of houses Height
DISCRETE RANDOM VARIABLES
Heres what a discrete random variable can look like:
1 2 2.5 3
Pr( = ) 0.1 0.2 0.3 0.4
From this we can deduce that:
0 Pr( = ) 1
Pr = = 1
DETERMINING PROBABILITIES
1 2 2.5 3
Pr( = ) 0.1 0.2 0.3 0.4
The probability of an event occurring can be established
straight from the table
Pr = 2.5 = 0.3
The probability that spans more than one discrete value can
be established through addition
Pr > 2 = Pr = 2.5 + Pr = 3 = 0.7
EXAMPLE
For the following discrete random variable, find all possible
values of and calculate Pr( 2) .
1 2 2.5 3
Pr( = ) 0.2 3 0.4
The sum of the probabilities must equal one
+ 2 + 0.2 + 0.4 = 1
= 0.1
Determine the probability by adding the required values
Pr 2 = Pr = 2 + Pr = 2.5 + Pr( = 3)
= 0.2 + 0.3 + 0.4 = 0.9
EXPECTED VALUE (MEAN)
Measures the centre of the distribution
E = = Pr( = )
Is calculated by summing the multiples of the random
variables and their respective probabilities
1 2 2.5 3
Pr( = ) 0.1 0.2 0.3 0.4
E = 1 0.1 + 2 0.2 + 2.5 0.3 + 3 0.4 = 2.45
VARIANCE
Measures the spread of the distribution
Var = 2 = E 2 E X 2
Is calculated by summing the squares of the variables and
their respective probabilities, less the square of the mean
1 2 2.5 3
Pr( = ) 0.1 0.2 0.3 0.4
E 2 = 12 0.1 + 22 0.2 + 2.52 0.3 + 32 0.4 = 6.375
Var = 6.375 2.452 = 0.3725
STANDARD DEVIATION
Is also a measure of spread of a distribution
= = Var() = E 2 E X 2
Is calculated by taking the square root of the variance
Notice this means that the variance can never be negative
= Var = 0.610
EXAMPLE
For the following discrete random variable find the mean and
the standard deviation.
2 0 2 4
Pr( = ) 0.1 0.3 0.1 0.5
E = 0.2 + 0.2 + 2 = 2
E 2 = 0.4 + 0.4 + 8 = 8.6
Var X = 8.6 22 = 4.6
= 4.6 = 2.144
EXPECTATION THEORIES
Expectation theories allow us to work with the mean and
variance of a random discrete distribution without re-writing
the distribution
E + = E +
Var + = 2 Var()
The variance has no horizontal translation, why do you think
this is?
EXAMPLE
2 0 2 4
Pr( = ) 0.1 0.3 0.1 0.5
E =2 and Var = 4.6
Given that = 3 + 2, Find Var , E() and its standard
deviation
E =32+2=8
Var = 32 4.6 = 41.4
= 41.4 = 6.434
APPLICATION
The amount of cars that pass through a certain drive-thru on
any particular day follows the discrete random variable, .
0 6 9 12
Pr( = ) 0.1 0.3 0.1 0.5
A. What is the average amount of cars that pass through?
B. On a particular day, a few cars have passed through the
drive-thru, what is the probability that the amount of cars is
greater than 6?
C. What is the probability that no cars pass, 5 days in a row?
APPLICATION (A)
0 6 9 12
Pr( = ) 0.1 0.3 0.1 0.5
What is the average amount of cars that pass through?
E = 6 0.3 + 9 0.1 + 12 0.5 = 8.7
It is expected that, on any given day, that 8.7 cars will pass
through the drive thru.
APPLICATION (B)
0 6 9 12
Pr( = ) 0.1 0.3 0.1 0.5
On a particular day, a few cars have passed through the drive-
thru, what is the probability that the amount of cars is greater
than 6?
This is a conditional probability question.
Pr( > 6) 0.6 2
Pr > 6 > 0 = = =
Pr( > 0) 0.9 3
APPLICATION (C)
0 6 9 12
Pr( = ) 0.1 0.3 0.1 0.5
What is the probability that no cars pass, 5 days in a row?
Each day is independent of each other, so we determine the
probability of it occurring on day one and raise it to the power
of 5.
Pr = 0 = 0.1
Pr 0,0,0,0,0 = 0.15 = 0.00001
EXAM TIPS
You must know all of the properties of probability before
heading into your first exam, without them you will not be
able to answer most questions
The terminology, and their respective formulae (union,
independence, mutual exclusivity, conditional, etc.) are the
same for all distributions.
Express your answers in the required form, especially when
dealing with expected value.
Discrete probability questions always appear in exam 1
SUMMARY
Integrations basic rules
Definition of a probability
Sample space
Terminology
Union, intersect
Probability properties
Independent, conditional and mutually exclusive
Discrete random variables
Mean, variance, standard deviation and
expectation theories
LEARNING OUTCOMES
Binomial Distributions
Factorial and combinations
Bernoulli sequences
Expectancy and Variance
Continuous Probability
Continuous random variables
Measures of centre
Measure of spread
BERNOULLI SEQUENCES
A Bernoulli sequence is a sequence of trials with the
following properties:
Each trial result has one of two outcomes, either
success or failure
The success of a trial, , is constant for all trials,
and each failure of a trial, 1 , is also constant
The trials are independent, such that any one trail
does not affect any other trial
An example could be a coin toss where heads is success,
as each event is independent and the chance for success
and failure is the same throughout all trials
BINOMIAL DISTRIBUTION
If , is the number of successes in a Bernoulli sequence,
each with a success probability of , then , is called a
binomial random variable, and has a binomial distribution.
We write any binomial random variable as follows:
~ ,
The difference between a binomial distribution and a
discrete one is that order doesnt matter.
For example, obtaining 5 favourable outcomes out of 10
trials, can all occur at the start or towards the end.
FACTORIAL AND COMBINATION
Factorial is another operator and is represented using the
exclamation mark such that:
! = 1 2 2 1
The combination formula represents the amount of groups
of size , that can be formed from entities
!
=
! !
EXAMPLE
Evaluate 6!.
6! = 6 5 4 3 2 1 = 720
How many different combination of 4 students can be chosen
from a group of 15?
15 15! 15 14 13 12
= = = 1365
4 4! 11! 4321
There are 1365 different combinations.
DERIVATATION OF FORMULA
If we were to observe trials and obtain successes, of
probability,, regardless of order, we would:
Pr 3 success, 2 fails = 1 1
Pr success, fails = 1
Ways we can arrange the amount of successes =
Pr = = 1
EXAMPLE
Given a soccer player, in practice, has a 90% chance of kicking
a goal, what is the probability that he makes all his shots if he
takes 10? What is the probability he makes half his shots?
~(10, 0.9)
10
Pr = 10 = 0.910 0.10
10
= 0.910
10
Pr( = 5) = 0.95 0.15
5
= 252 0.95 0.15
= 0.001 (3d. p)
EXAMPLE CONTINUED
What is the probability that he makes at least 1 of his shots?
~(10, 0.9)
In scenarios like this, you can add all the probabilities from 1
to 10, however, as the sum of all probabilities must total 1, we
can take the complement and subtract it from the total
probability.
Pr 0 = 1 Pr = 0 = 1 0.110 = 0.999
EXPECTANCY AND VARIANCE
Expected value, or mean, is the average amount of
successes we can expect given our distribution
E =
Variance measures the spread of the distribution
Var = (1 )
EXAMPLE
Stephen Curry makes 55% of his three point shots in a typical
game. Given that he takes 16 shots in a particular game what
is the expected amount of shots he will make? What is the
variance of Currys shooting distribution?
~(16, 0.55)
E = = 0.55 16 = 8.8
Var = 1 = 0.55 16 0.45 = 3.96
CALCULATORS
Being able to utilise a calculator is extremely important. It is
integral, sorry, as we most of probability is on exam 2.
Speed and efficiency is the most important.
The following information will be for the TI-nspire CX CAS
For all other calculators, make sure you know shortcuts to
bring up the screens required. Usually they are titled
Binomial CDF or Binomial PDF.
EXAMPLE CONTINUED
On game 5 of the NBA finals, Curry makes a few shots, what is
the probability that he made two of his shots, given that he
shot the ball 9 times?
~(9, 0.55)
Pr( = 2)
Pr = 2 > 0 =
Pr( > 0)
0.041
=
0.999
= 0.041
Curry has a 4.1% chance to make 2 of his 9 shots.
CONTINOUS PROBABILITY
A continuous variable is one that can take any real number
on a number line
A continuous random variable has the following properties:
0 for all
The area enclosed by the graph and the
must equal to one
Which of the following are continuous variables?
Money Gender Postcodes
Months in progress Number of houses Height
EXAMPLE
Is the follow a continuous random variable?
3 2
= 01
0 elsewhere
Is () 0 for all ?
3 2 increases as does, 0 = 0 0 for all
Is the area under the graph 1?
1
= 1
0
DETERMINING PROBABILTIIES
Probabilities for continuous functions only occur over
intervals, that is we cannot have Pr( = ) .
We determine probabilities using integration
Pr < < =
Whether or not the signs are inclusive is irrelevant, we
obtain the same answer
EXAMPLE
2
=3 01
0 elsewhere
Pr 0 < < 1 =
Is a C. R. V. so Pr 0 < < 1 = 1
1
Pr 0 < < =
2
1/2
3 2 = 0.125
0
1
Pr <<1 =
2
1
1 Pr 0 < < = 0.875
2
EXPECTANCY
The expected value, or mean, of a continuous random
variable, , is given by:
= E =
If () is a function of then the expected value of () is
given by:
E = ()
MEDIAN
The median is another measure of centre and is the middle
most value such that 50% of the distribution lies above the
median, and 50% below.
The median, , is calculated as follows:
= 0.5
In general, the median is different to the mean.
VARIANCE AND STANDARD DEVIATION
Variance and standard deviation are measures of spread
Variance is calculated as follows
Var = E X 2 2
E X 2 = 2
Standard deviation is the square root of variance
= V(X)
INTERQUARTILE RANGE
The interquartile range is the difference between the 75th
percentile and the 25th percentile, that is, it is the range in
which the middle 50% of values exist.
IQR =
= 0.25 and = 0.75
EXAMPLE
Consider the following probability density function:
+ 2 , 01
=
0,
2
If E X = find and .
3
1 1
2
+ 2 = 1 and + 2 =
0 0 3
1
= and = 2
3
EXAMPLE CONTINUED
1 2,
= 3 + 2 01
0,
Evaluate Var(X)
Var X = E X 2 2
1 2
1 2
= 2 2
+ 2
0 3 3
23 4
=
45 9
1
=
15
EXAMPLE CONTINUED
1 2,
= 3 + 2 01
0,
Find the median for the distribution.
1
+ 2 2 = 0.5
0 3
25 3
+ = 0.5
5 9
= 0.995
The median is 0.995, correct to 3 decimal places.
EXAMPLE CONTINUED
1 2
= 3 + 2 , 01
0,
Find the interquartile range for this distribution.
= 0.25 and = 0.75
= 0.5 and = 0.881
IQR = = 0.381
EXAMPLE CONTINUED
1 2,
= 3 + 2 01
0,
Find Pr( > 0.2| < 0.7)
Pr(0.2 < < 0.7)
Pr( > 0.2 < 0.7 =
Pr( < 0.7)
0.39
=
0.462
= 0.844
EXAM TIPS
Binomial probability rarely appears in exam 1, if it does it
will be scenarios that involve small numbers.
Students often have trouble using the new screens on their
calculator, get practice on it, speed test yourself, this is
what separates students.
Rounding is also an issue, students are careless when
answering questions, round your answers to correct decimal
places.
Note in exam 1, if your answer is incomputable or is difficult
to compute, you most likely messed up.
EXAM TIPS
Continuous probability rarely appears in exam 1, if it does it
would be a simple integration question, or the final question
on the exam, such as in 2016.
That means, get to know your calculator, speed and
accuracy is what sets students apart.
This is the easiest of the all the probabilities as no new
knowledge is required to compute problems, no new
screens to worry about. This should be the quickest section
in your exam.
SUMMARY
Binomial Distributions
Properties of a binomial distribution
Determining probabilities
Expected value and variance
Understanding the binomial distribution formula
Using your calculator
Continuous distributions
Properties of a P.D.F
Determining probabilities
Mean, median, variance, standard deviation and
IQR
LEARNING OUTCOMES
Normal Distributions
Standardisation and the 68 95 99.7% rule
Determining normal probabilities
Solving problems
Central limit theorem
Statistics
Populations and samples
Large and small populations
Distribution of sample proportion
Confidence intervals
NORMAL DISTRIBUTION
When something is normally, or approximately, normally
distributed, it means the data is centred around a mean
and follows a bell curve.
The normal distribution is a special type of continuous
probability distribution and is written like:
~(, )
1 1 2
= 2
2
THE BELL CURVE
Bertrand,. (2016, 20/5). Miller Analogies Test Bell Curve. https://magoosh.com/mat/miller-analogies-test-bell-curve.
BELL CURVE EXPLAINED
Most data lies around the centre
The further out you go, the less data
Each line represents a standard deviation from the mean,
the more standard deviations the further out
68% of the data lie within 1 standard deviation of the mean
95% within 2
99.7% within 3
68 95 99.7% RULE
If the age of students in a particular awesomely spectacular
methods lecture was normally distributed with a mean of 17
and a standard deviation of 0.3:
Over what range, do we expect 68% of the students to lie?
~(17, 0.32 )
Using our rule, 68% of data lies within 1 standard deviation of
the mean:
16.7 and 17.3
EXAMPLE CONTINUED
~(17, 0.32 )
Using the rule, find such that Pr < < 17 = 0.475
We have discovered that the value for is two standard
deviations below the mean
= 17 2 0.3 = 16.4
THE STANDARD NORMAL
The bell curve you just saw was one of many, it was the
standard normal distribution and has special properties
~(0, 1)
We can standardise any normal distribution:
The standardised value tells us the number of standard
deviations the value lies from the mean
EXAMPLE
~(17, 0.32 )
How many standard deviations from the mean is the age 17.6?
17.6 17
= =2
0.3
How many standard deviations from the mean is the age 16.7?
16.7 17
= = 1, =1
0.3
Hence evaluate Pr(16.7 < < 17.6) using the rule.
0.68 0.95
Pr 16.7 < < 17.6 = + = 0.815
2 2
EXACT VALUES
The rule is only approximate, for example, not exactly 68%
of the data lies between 1 standard deviation of the mean,
but its close enough
In order to determine the exact values we will need
assistance from our calculators
We will use the Normal CDF function on our calculators
for this purpose
EXAMPLE
Find the exact percentage of the amount of data that lies
within one standard deviation of the mean.
normCdf(1, 1, 0, 1)
= 0.682689
This was performed on a TI - nspire CX CAS
Notice that it is not exactly 68%, this is true for the other
percentages as well.
EXAMPLE
The amounts of spins per minute on a fidget spinner is
normally distributed with a mean of 120 and a standard
deviation of 10. Find the probability that the amount of spins
is greater than 117.
~(120, 102 )
normCdf(117, , 120, 10)
= 0.6179
INVERSE NORMAL
The inverse normally, as the name may suggest, is the
opposite of the normalCdf function.
When we used the normalCdf before we plugged in the
lower and upper bound to give us the area
The inverse normal we can provide the area and it will give
us the upper bound, ASSUMING, that the lower bound is
.
This is for the TI - nspire CX CAS, some calculators may
allow you to set both the lower and upper bound
EXAMPLE
The amounts of spins per minute on a fidget spinner is
normally distributed with a mean of 120 and a standard
deviation of 10.
It is found that, 10% of fidget spinners are defective. A spinner
is defective is it more than amount of spins. Find . (3dp)
Pr > = 0.1
Pr < = 0.9
invNorm(0.9, 120, 10)
= 132.8155
= 132.816
DETERMINING AND
Sometimes we are not given the standard deviation or the
mean, however we can determine these using the inverse
normal function. Recall that:
By standardising any normal distribution we can use the
inverse function as we know what the mean and standard
deviation is
~(0, 1)
EXAMPLE
is normally distributed with a mean and standard
deviation . It is known that 35% of the data lies under = 6.
And 45% lies above = 10. Find and .
Pr < = 0.35 Pr > = 0.45
6 10
Pr < = 0.35 Pr > = 0.45
6 10
Pr < = 0.35 Pr < = 0.55
invNorm(0.35 , 0, 1) invNorm(0.55, 0, 1)
6 10
= 0.385 = 0.126
= 9.016 = 7.828
STATISTICS
Statistics is taking all about taking a sample from a
population
In the real world, the population is too large, so we work
with statistics
A population can be defined as anything, for example, this
lecture room can be a population and the people at the
front are a sample.
SAMPLE PROPORTION
A sample proportion is the ratio of desired outcomes to
total outcomes in a certain sample
For example, if I was interested in red balls, and I take a
sample of 10 balls, where in which 5 of them were red, then
1
my sample proportion is .
2
The sample proportion is denoted with .
The distribution of sample proportion is .
SMALL POPULATION
Sample proportions are all based on the binomial
distribution
A small population, is a population of size about equal or
less to approximately 20.
When establishing the distribution of sample proportions
we use counting methods.
EXAMPLE
In a population of 10 batteries, it is known that 6 of these are
defective. Determine the sample proportion distribution of
defective batteries if I take a sample of 3.
3 6 4 3 3
Pr = 1 = =
1 10 9 8 10
0 1 2 3
Pr( = ) 1 3 1 1
6 10 2 30
0 1 2 1
3 3
Pr( = ) 1 3 1 1
6 10 2 30
LARGE POPULATION
A large population has an uncountable amount of entities,
so instead of using counting methods we must use the
binomial distribution
For a large sample you will be given a population
proportion, . This will be our success rate.
We will still take a sample size to create a sample propotion
distribution
EXAMPLE
A shipment of ten million fidget spinners has arrived. It is
known that 20% of all fidget spinners are red. We take a
sample of 100 spinners.
Evaluate Pr > 0.1
Pr > 0.1 = Pr > 10
= 0.994
EXAMPLE CONTINUED
Evaluate Pr( 0.2| < 0.3)
Pr(20 < 30)
Pr 0.2 < 0.3 =
Pr( < 30)
0.5398
=
0.9939
= 0.5431
CENTRAL LIMIT THEOREM
When we have a large population the distribution can be
approximately normal
Which means we have methods of determining the
standard deviation and mean
= =
(1 )
=
CONFIDENCE INTERVAL
A confidence interval is determined from a sample
It is the estimate, such that a certain percentage of all
sample proportions will lie within this confidence interval
In general the confidence interval is:
1 1
, +
Where Pr < < =
100
EXAMPLE
A certain sample of 100 Spotify users were taken, out of these
80 were premium users. Determine a 95% confidence interval
for .
80
= = 1.96
100
80 0.8 0.2 80 0.8 0.2
: 1.96 , + 1.96
100 100 100 100
451 549
: ,
625 625
SUMMARY
Normal Distributions
68 95 99.7% Rule
Determining exact and approximate areas
Determining inverse normals
Using your calculator
Statistics
Sampling from small and large populations
Distributions of sample proportions
Expected value and standard deviation
Confidence intervals
Questions?
