AID-521 Mathematics for Data Science
Module: Probability | Lecture: 1
BASIC CONCEPTS OF
PROBABILITY
random events, probability calculation
Basic Definitions
A random event is one of the outcomes of an
underlying process, whose outcome at a particular
time is not known prior to that time.
→ On Monday, that the Sun will rise in the east is a
deterministic event.
→ On Monday, that the stock index will
fall below 100 points is a random event.
The underlying process is called experiment.
2
Basic Definitions
A trial of a process is the (deliberate/natural)
performance of the process once, after which its
outcome can be observed.
→ A medical test is conducted on a patient to infer
the effectiveness of a drug.
→ The duration of rainfall in Delhi can be
shorter or longer.
Multiple trials of a given process may or may not be performed.
3
Basic Definitions
The sample space S of a trial is the set of all possible
outcomes of the trial.
→ In a coin toss, the sample space is composed of
heads and tails.
→ The sample space of the price of a new car
contains all non-negative real numbers, i.e., [0, ∞).
So, a random event A is a subset of S. Can you see this?
4
Defining Probability
Intuitive Definition
The probability of an outcome (event) is the
proportion of times the outcome (event) would occur
in a long run of repeated experiments.
→ Requirement of identical conditions for repeated
experiments.
→ Unbiased vs. Biased coin: P(H) = limn→∞ #Hn
?
5
Defining Probability
Axiomatic Definition*
Given a sample space S composed of events,
probability P is a function that satisfies the following
conditions.
→ P : S → [0, ∞), i.e., P(A ⊂ S) ≥ 0 for all events in S
→ P(S) = 1
→ P(A ∪ B) = P(A) + P(B), if events A and B are
mutually exclusive (i.e., A ∩ B = ϕ)
*Does not say how to calculate/assign probabilty to an event. If a
function P satisfies the above conditions, it is a probability
function.
6
Properties of Probability
→ P(Ac ) = 1 − P(A)
→ If A ⊂ B, then P(A) ≤ P(B)
→ P(A ∪ B) = P(A) + P(B) − P(A ∩ B)
for any two events A and B in S
7
Properties of Probability
Try yourself
In a large university, the freshman profile for one
year’s fall admission says that 40% of the students
were in the top 10% of their high school class, and that
65% are white, of whom 25% were in the top 10% of
their high school class. What is the probability that a
freshman student selected randomly from this class
either was in the top 10% of his or her high school
class or is white?
8
Properties of Probability
Try yourself – solution
→ Let A be the event that a person chosen at random
was in the top 10% of his or her high school class,
and let B be the event that the student is white.
→ We have P(A) = 0.4, P(B) = 0.65, P(A ∩ B) = 0.25
→ Required Event1 is A ∪ B
→ Required Probability = 0.4 + 0.65 − 0.25
1
student chosen is white or was in the top 10% of his or her high
school class
9
Counting: Ordered Objects
From n objects, we are sampling m items:
with replacement
total no. of ways
= n ways × n ways × · ·· for m times
= nm
without replacement
total no. of ways
= n ways × (n − 1) ways × · · · × (n − m + 1) ways
n!
= (n−m)! := nPm
10
Counting: - Un-ordered* Objects
*A sample ”ABC” is the same as ”BAC” – i.e., they are not two different samples.
From n objects, we are sampling m items:
without replacement
total no. of ways
= n ways × n ways × · ·· for
( n )m times, divided by m × · · · × 1
n!
= (n−m)! × m!
1
:= nCm or m
with replacement
total no. of ways
= equivalent to ”choosing” m samples from n + m − 1
objects without replacement
= n+m−1Cm
11
Counting: Objects in Classes
No. of ways to group n objects in m classes
with
∑m ni objects in the i-th class (i = 1, 2, ..., m and
i=1 ni = n)
( )
n
=
n1 · n 2 · · · n m
Hint: how many samples are you choosing from the n objects?
n!
=
n1 ! · n2 ! · · · nm !
12
Counting & Probability
In the previous slides, we saw how to count cases.
But, where is probability?
Suppose there are N possible outcomes of an
experiment, and
let nA be the number of outcomes in an event A.
Then the probability of event A is
nA
P(A) = .
N
13
Counting & Probability
Try yourself
The admissions committee of a department at a U.S.
university is selecting students.
Suppose that the admission committee decides to
randomly choose seven graduate students from a
pool of 30 applicants, of whom 20 are foreign and 10
are U.S. applicants.
What is the probability that a chosen seven will have
four foreign students and three U.S. students?
14
Counting & Probability
Try yourself – solution
(30)
No. of ways of selecting 7 applicants out of 30 = 7
No.
(20of
) ways
(10)of selecting 4 foreign and 3 U.S. students
= 4 × 3
Hence, required probability
(20) ( )
× 10
= 4
(30) 3 = · · · = 0.286
7
15
IF THIS LECTURE WAS HEAVY FOR YOU...
It is strongly recommended to go through
Ramachandran and Tsokos, and to do the solved
examples yourself.
