Probability
• Modern statistics also has to provide for
estimation of population parameters on
the basis of sample statistic.
• Drawing inferences about sample
statistic on the basis of population
parameters.
• Testing of hypothesis with regard to 1
and 2 above.
• Decision making under risk and
uncertainty by estimating the degree of
risk.
• A numerical measure of uncertainity is
provided by a very important branch of
Statistics called the “Theory of Probability”
• Prof Ya-Lin-Chou “Statistics is the science
of decision making with calculated risks
in the face of uncertainity.”
Terminology
• Experiment
• Any operation that results in two or more
outcomes is called an experiment.
• Eg.Tossing a fair coin is an experiment.
• Rolling an unbiased die .
• Drawing a card from pack of cards.
• The result of a random experiment is
called outcome.
• The performance of a random experiment
is called a trial.
• The outcome is an event
• Event ---- Simple or Compound
• Simple Event : if it corresponds to a single
possible outcome.
• Eg.Tossing a die–
• chance of getting 3
• chance of getting an odd number.
Favourable Cases
• The number of outcomes which result
in the happening of a desired event are
called favourable cases.
• Eg.single throw of a die---- the number
of favourable cases for getting an odd
number are three.ie 1,3 and 5 .
Mutually Exclusive Events
• Two or more events are mutually
exclusive if the happening of any one of
them prevents the happening of all other
in a simple experiment.
• Eg.Tossing a coin---getting head and
tail .
• If a baby is male, cannot be female
Equally Likely Events
• Two or more events are said to be
equally likely if the chance of their
happening is equal.
• Eg.Throwing of an unbiased dice, coming
up of 1,2,3,4,5 or 6 is equally likely.
Independent Events
– Two events are independents if the occurrence of a
event does not affect the occurrence of the second
event.
– Example
• If the first new born is male, does not affect that
the next be female.
– Probability of two independent events is obtained
multiplying individual probabilities of each event.
– This is the multiplicative law of probability.
Definitions
• Probability is the possibility that an event
occur.
• It measures the likelihood that a particular
event will or will not occur.
• Probability of an event=No. of favourable
cases for the event/Total no. of cases both
favourable as well unfavourable to the event
•
Examples
• If we throw a coin in the air once, the
probability to obtain head is ½, because
only once we can obtain head or tail.
• Suppose in a class there are 100 students
out of whom 90 boys and 10 are
girls.probability of a student chosen at
random out of the student in the class to
be a girl is 10/100=0.1
Modern approach to Probability
• Introduces probability as a number
associated with each event.
• Based on certain axioms which express the
rules for operating with such numbers.
• This means that the probability of events
must only satisfy these axioms.
Axiomatic Approach to
Probability
• Let S be the sample space
corresponding to a random experiment
and let A be any event.Then P(A) is
known as the probability of an event A
if it satisfies the following axioms.
• Axiom 1 : P(A) >= o
• Axiom 2 : P(S) = 1
• Axiom 3 : If A and B are disjoint events
then P(AUB)= P(A)+P(B).
• If a result is sure that occur the probability
will be 1 (100%).
• Probability of an event happening is
denoted by “p”
• If a event is sure that does not occur the
probability will be 0.
• Probability of an event not happening is
denoted by ‘q’
Probability
• Probability to have
amebiasis is 59/200=
0.295= 29.5% Results for n (%)
E. histolytic
Positive 59 (29.5)
• Probability of does not Negative 151 (70.5)
have amebiasis es
151/200= 0.705 = 70.5% or
1 - 0.0.295=0.705 = 70.5%
Empirical Distributions
• Observed frequency distributions -
empirical distributions or experimental
distributions
• Used to estimate certain values in the
population on the basis of sample studies.
• Situations where we can derive expected
values on the basis of some relationships.
• Eg.Suppose we toss a coin.
• Probability of a head is ½ and tail is ½.
• Toss a coin 100 times expected no. of heads is
50
• .
• Toss a coin 100 times.Heads come up 60
times.This is the observed frequency of heads,
Theoretical Frequency
Distributions
• Theoretical or probability distributions are
such distributions which are not obtained
by actual observations or experiments but
are mathematically deduced on certain
assumptions.
Standard probability Distributions
• Binomial Distributions
• Poisson Distributions
• Normal Distributions OR Gaussian
Distributions
Height in cms No of men
142.5 3
145.0 8
147.5 15
150.0 45
152.5 90
155.0 155
157.5 194
160.0 Mean 195 ( 68 % )
162.5 136
165.0 93
167.5 42
170.0 16
172.5 6
175.0 2
Normal Curve
• It is often called the corner stone of
statistical analysis and drawing of
inferences.
Properties of Normal Curve
• It is a bell shaped curve
• It is symmetrical about its mean
• At the centre of the distribution which is
peaked all the three measures of central
tendency mean,median and mode
coincide.
• Total area of the curve above the x-axis is
one sq unit
• It has two points of inflections(points at
which the curve changes its direction) are
at a distance of one standard deviation
from the mean
• About 68.268% of the observations will be
falling within mean plus or minus one
standard deviation,95.45% within two
standard deviations and 99.73% within
three standard deviations
• It is mesokurtic.
• The curve is asymptotic to the base line.It
continues to approach but never touches
the base line.
• It is unimodal as there is only one point of
maximum frequency.
Importance of the normal
distribution
• The whole data can be summarised by two
quantities – mean and sd
• Can find out the probability of getting a
person with a specific characteristic in a
group
• Can be used to find confidence limits of
population parameters
Variability and AveragesPatients
Controls
• Graph 1: Bipolar
Frequency
disorder
– Different variability
– Same averages
Depressed Manic
Patients
Frequency Controls
• Graph 2: Blood sugar
levels
– Same variability
– Different averages
Low High
The Normal Distribution
• Normal distributions with different
population variances and the same
population mean
2 = 1
2 = 2
f(x) 2 = 3
2 = 4
The Normal Distribution
• Normal distributions with different
population variances and different
population means
2 = 1 1
2 = 3 3
Standard Normal Curve
• Normal curve in which mean is zero and
the standard deviation is unity.
• Z,standard normal variate
• Area under the standard normal curve is
unity.
Standard Normal Variate
• Deviation of an individual observation
from the mean in a normal distribution or
curve is called standard normal variate
and is given the symbol Z.
• It is measured in terms of standard
deviation (SD) and indicates how such an
observation is bigger or smaller than mean
in units of standard deviation.
• Z = Individual observation - Mean/SD
• Z is called the standard normal deviate or
relative normal deviate
• Eg: If x=170 cm , mean = 166 cm and
• SD = 4cm then
• Z = 170 – 166/4= 1
"Life is good for only two
things, discovering
mathematics and teaching
mathematics"
Siméon Poisson (1781-1840)
