PROBABILITY

PROBABILITY
 Random Experiment or Trial (R.E) : the occurence of any one of them prevents (or)
A repeatable experiment the results of which are eliminates the occurence of all the other events of
known in advance but exact result will be known that R.E.
only after the experiment is completed is known Two events A and B are said to be M.E. or disjoint
as random experiment or an experiment, result of when both cannot happen simultaneously in a single
which can not be predicted with certainity is called trial or experiment. i.e., A  B   .
random experiment. Ex: In tossing a coin head turning upwards eliminates
a) Tossing a coin tail turning upwards in that particular R.E. So getting
b) Rolling a die head & tail are M.E. events.
Drawing a specified number of cards from a well
The events A1, A2 , A3 ,....., An of a R.E. are said to be
shuffled pack of 52 playing cards etc.,
 Sample Space (S) : The list of all possible M.E. or disjoint events if, Ai  A j   , for i  j &
outcomes or results of a random experiment is 1  i, j  n .
called as sample space for that experiment and it Note : Whenever it is possible for two or more
is denoted by S. number of events to happen simultaneously, then
Ex: i) If a coin is tossed once, the sample space S is those events are said to be compatible.
given by S = {H,T}  Equally likely events (E.L.E.): The events
ii) If 2 coins are tossed once, the sample space of a R.E. are said to be E.L. When one does not
S={HH,HT,TH,TT} occur more number of times than any one of the
iii) If a die is rolled once the sample space other events i.e., if all the events occur same number
S={1,2,3,4,5,6} of times.
 Elementary event or Indecomposable The events of a R.E. are said to be equally likely
event: The possible outcome or the result of a when there is no reason to expect a particular event
random experiment that can not be split further is in preference to any other event.
called an elementary or simple event. Ex: If an unbiased coin is tossed a large number of
 Event: An outcome or the result of a times, each face i.e., head and tail can be expected
random experiment is called an event. to appear same number of times.
Ex: While tossing a coin we may get head or tail  Exhaustive events (E.E.): The events of a
upwards. Now getting head upwards is an event random experiment are said to be exhaustive, if
or getting tail upwards is also an event. the occurence of any one of them is a certainity.
Ex: The event of getting head upwards is an elementary A set of events is said to be exhaustive, if these
event. include all possible outcomes of the R.E.
 Composite event : The possible outcome or Ex: In tossing a coin it is certain that either the head or
the result of a random experiment that can be further the tail will turn upwards. So head & tail put together
split into more than one elementary event is called are called as exhaustive events.
a composite event.  Complementary Events : Suppose A is any
The event can be either an elementary event or a event of a random experiment associated with
composite event. sample space S.
Ex: Getting a face with odd number of points upwards The complementary event of an event A denoted
in rolling of a die is a composite event, since it can
by Aor AC is the event given by AC   S  A 
be divided further into three elementary events
namely: Ex: If two coins are tossed, S={HH,HT,TH,TT}
 Mutually exclusive event (M.E.E.): Let A: occurence of atleast one head
The events of a R.E. are said to be M.E. events, if Here AC is the event of non - occurrence of atleast

1
 PROBABILITY

From these we can conclude that the probability

one head i.e, AC is the event of getting both tails. of any event is a non-negative rational number
Here A={HH,HT,TH}, AC ={TT} which lies between 0 & 1, both inclusive.
Note : Let S be a sample space. The event  is
called impossible event and the event S is called  A non-leap year = 52 weeks + 1 day
certain event in S. A leap - year = 52 weeks + 2 consecutive days of
 Favourable Cases: The possible outcomes or a week.
results of a R.E. which leads to the occurence of
the event A are called as favourable cases for the  Description of Pack of Cards:
event A to happen.
Ex: In rolling a symmetrical die 1, 3 & 5 are happening Pack
for the event A, the event of getting odd number of of 52
cards
points upwards. So they are called as favourable
cases for the event A.
 Classical Definition of Probability (or) Black Cards Colours Red Cards
Mathematical Definition of Probability
(or) A priori Probability: The probability of
the event A, denoted by P(A) is defined as (13) (13) (13) (13)
Spade Club Heart Diamond
m n  A    
P  A  
n nS  .
 Here clubs, spades, diamonds, hearts are four suits.
Where n = The total number of possible outcomes
 Each suit contains 13 cards namely 2, 3, 4, 5, 6, 7,
in a R.E. which are M.E., E.L. & collectively
8, 9, 10, J, Q, K, A.
exhaustive, m = The number of possible outcomes
 In every suit J, Q, K are face cards (or court cards)
favourable to A in that R.E., n(S) = The total number
 In every suit J, Q, K, A are called honour cards.
of sample points in a sample space S, n(A) = The
 Jack card (J) is also called knave.
number of sample points favourable to A in that
 The cards bearing the digits 2,3,4,5,6,7,8,9,10 are
sample space S.
known as number cards. Ace is treated as 1.
m
p(A) = can also be expressed as "The odds in  Each suit contains 9 numbered cards.
n  In a pack of cards the number of cards with same
m face value is 4.
favour of the event A are m to n - m or " or
nm  Odds in favour of A : It is defined as P(A) : P(A)
"The odds against the event A are n-m to m or
Odds against A : It is defined as P(A) : P(A)
nm
". But in both the cases If P(A) : P(A) = x : y then
m

P  A 
favourable cases m m x y
total no. of cases
 
mnm n P(A) = and P(A) 
xy xy
m  Limitations:
Now consider P  A  
n If n is very very large, we can not determine the
If m = 0, P(A) = 0, then A is said to be an impossible probability of the event A with the help of this
event. definition.
n The probability can not be found when the
If m=n, P  A    1 , then A is said to be a sure or
outcomes are not equally likely.
n
certain event. When the outcomes are not M.E., our logic may
go wrong.
 The range of P(A) is [0, 1] i.e., 0  P  A   1
One of the serious drawbacks of this definition is
that in defining probability we use the term equally
2
 PROBABILITY

likely i.e., equally probable i.e., probability itself.

 All the identical trials are to be conducted
independently. Again to define the concept of
independence, we require the concept of
probability.
 Axiomatic approach to Probability :
Let S be a finite sample space associated with a
random experiment. Then a real valued function P
from the power set of S i.e., P(S) into the real line
R is called a probability function on S, if P satisfies
the following three axioms. happen
 Axiom of Positivity: For every subset A of S, or B does not happen
P  A  0 i.e., P  A  0, A  S
 Axiom of Certainty : P(S) = 1
 Axiom of Union or Axiom of Additivity:
 If A1 , A2 are two disjoint subsets of S then
P  A1  A2   P  A1   P  A2  .
Now the image P(A) of A is called the probability
of the event A.  Addition Theorem : If A and B are any two
Note: In general, if A1 , A2 , A3 ,....., An are ‘n’ disjoint compatible events in a sample space S, then the
subsets of S then P  A1  A2  A3  ......  An  probability of occurence of either the event A or B
or both the events A & B i.e., the event of A  B is
 P  A1   P  A2   P  A3   ....  P  An 
given by
 n
 n
P  A  B   P  A  P  B   P  A  B  .......... 1
i.e., P U Ai   P  Ai 
 i 1  i 1
If A and B are M.E. events, P  A  B   0 , then
 If S consists of ‘n’ equally likely elementary events
P  A  B   P  A   P  B  ..........  2 
and Ai is one such elementary event of S, then
From equations (1) & (2) we can conclude that
1
P  Ai   ; i  1, 2,3,......n . P  A  B   P  A  P  B 
n
 Notation : Let A and B be two events in a sample Further A , B  A  B
space S. P  A or P  B   P  A  B   P  A  P  B 
 If A and B are two events such that A  B, then
P(A)  P(B)
 If A, B are two events of S then
P(A  B)  P(A)  P(A  B)  P(A)  P(B)
 P(exactly one of A, B occurs)

  
= P A B  P A B 
= P  A  P B   2P  A  B 
 P  A  B  P  A  B

 PA  B  P A  B

3
 PROBABILITY

 If A , B are two events then

 Multiplication Theorem (or) Product
i) P  A  B   1  P  A  B 
Theorem (or) Theorem of compound
ii) P  A  B   1  P  A  B  Probability: If A and B are two events in a
  A B B  A B &  A  B  A   A  B sample space S such that P  A   0 & P  B   0,
 For three events A, B and C then the probability of simultaneous occurence of
P  A  B  C   P  A  P  B   P  C   P  A  B 
the two events A and B is given by
P  A  B   P  A  .P  B / A  P  B  .P  A / B  .
P  A  C   P  B  C   P  A  B  C 
 If A, B and C are M.E. events, the probabilities of  For three events A,B and C P  A  B  C 
compound events will be zero, then  P  A . P  B / A . P C / A  B 
P  A  B  C   P  A  P  B   P  C 
 For four events A, B, C and D P  A  B  C  D 
 If A, B and C are three events then
 P  A  .P  B / A .P  C / A  B  .P  D / A  B  C 
P(exactly one of A, B, C occurs) =
P  A  P  B   P  C   2 P  A  B 
 Independent Events : Two events A and B of
2 P  A  C   2 P  B  C   3P  A  B  C  an experiment are said to be independent if
 P(exactly two of A, B, C occur) = P  A  B  occurence of A cannot influence the happening of
the event B
 P  A  C   P  B  C   3P  A  B  C 
If A & B are independent then P  B / A  P  B  &
 P(at least two of A, B, C occurs) = P  A  B 
P  A / B   P  A .
 P  A  C   P  B  C   2P  A  B  C 
 Two events A and B are independent if and only if
 If A,B and C are mutually exclusive and exhaustive
events of the sample space S, then A  B  C  S P  A  B   P( A).P( B)
and P(A)+P(B)+P(C)=P(S)=1  If A, B and C are independent, then
P A  B  C  P  A .P  B .P C 
 Conditional Event: If A,B are two events in a
sample space then the event of happening of B  If A, B, C and D are independent, then
after the happening of event A is called conditional P  A  B  C  D   P  A  .P  B  .P  C  .P  D 
event . It is denoted by B/A
Note: If A & B are independent,
 Conditional Probability: If A & B are two
P  A  B   P  A P  B  then
events in a sample space S such that P  A  0 ,
then the probability of occurence of B, after the P  A  B   P  A   P  B   P  A .P  B  . (or)
occurence of the event A is called the conditional
P  A  B   1  P  A  .P  B  .
probability of B given A, denoted by either P  B / A
If A, B & C are three independent events, then
B
or P  A  or P  B; A and is defined as P  A  B  C   1  P  A  .P  B  . P  C 
 
 If two events A & B are independent
P  A  B n A  B
P  B / A  
P  A n  A a) A & B are independent.
Similarly if P  B   0, then b) A & B are independent.

P  A  B n  A  B c) A & B are independent.

P  A / B  
P  B n  B  Three events A, B & C are said to be mutually
independent if
In general P  B / A  P  A / B 
4
 PROBABILITY

a) P  A  B   P  A .P  B 
 Baye’s Theorem: Suppose E1, E2 , E3 ,......, En
b) P  A  C   P  A .P  C  are 'n' mutually exclusive and exhaustive events in
c) P  B  C   P  B  .P  C  sample space S with P ( Ei )  0 for i  1,2,3,....., n
in a random experiment .Then for any event A of
d) P  A  B  C   P  A .P  B  .P  C  the random experiment,
Suppose the 1st three conditions (i), (ii) & (iii) are
satisfied and the last i.e., (iv) condition is not P ( Ek ).P  A 
P  k   n
E  Ek 
satisfied, then the three events A, B and C are said  A  for i=1,2,3,...,n
to be pair-wise independent.  P ( Ei ) P  A 
i 1  Ei 
 For any two independent events A and B
1
W.E.12:- A man is known to speak the truth 2 out
P (A  B)  (A  B)  of 3 times. He throws a die and reports that it
4
is a six. Then the probability that it is actually
 If A and B are two events, then a six is
max 0,1,..., P(A)  P(B)  P(A  B) Sol: Let A1 be the event that six occurs on the die and

 Min { P(A), P(B) } A2 be the event that a digit other than six occurs.
Let E be the event that the man makes a statement
that six is obtained.
 Law of Total Probability : Let in a random
experiment S is a sample space and P  E / A1  = prob. that man speaks truth= 2/3
E1 , E2 , E3 ,...., En are mutually exclusive and
P  E / A2  = probability that the man does not
exhaustive events. If A is any event which occur
with E1 or E2 or E3 or .... or En , then speak truth = 1/3
By Baye’s theorem,
 A  A
P  A   P  E1  .P    P  E2  P    ... P  A1  P  E / A1 
 E1   E2  P  A1 / E  = P A P E / A  P A P E / A
 1  1  2  2
 A
.....  P  En  P  
 En  2 1
.
3 6 2
W.E.11:- An urn A contains 3 white and 5 black 
balls. Another urn B contains 6 white and 8 = 2 1 1 5 7
.  .
black balls. A ball is picked from A at random 3 6 3 6
and then transferred to B.Then a ball is picked
at random from B. The probability that it is a  Geometric Probability :
white ball is (EAMCET 2010) Classical definition of probability fails if the total
Sol: Let A1 , A2 be the events of drawing white, black number of outcomes of an experiment is infinite.
balls respectively from bag A then Then the probability that a point selected in a given
P  A1   3 / 8, P  A2   5 / 8 region will be in a specified part of it is called
Let E be the event of drawing a white ball from geometrical probability or probability in continuous.
bag B after a ball is transfered from bag A to B. Thus the probability P is given by
Then P  E / A1   7 /15, P  E / A2   6 /15 Measure of specified part of the region
P
 P  E   P  A1  P  E / A1   P  A2  P  E / A2  Measure of the whole region
Where measure refers to the length, area, volume
3 7 5 6 17
=     of the region when we are dealing with one, two
8 15 8 15 40 or three dimensional space respectively.

5
 PROBABILITY
NOTE:  Two persons game : If 'p' and 'q' are the
 When 'n' fair coins are tossed the probability of probability of success, failure of a game in which A
n
cr and B play and if A starts the game then,
getting exactly 'r' ( n) heads (or tails) =
2n p 1
 When 'n' fair coins are tossed the probability of i ) probability of A’s win = 2 =
1 q 1 q
1
getting atleast one head (or tail) = 1  n qp q
2 ii) probability of B’s win = 2 =
 A coin is tossed (m+n) times (m>n), then the 1 q 1 q
probability of getting at least 'm' consecutive heads  Three persons game : If 'p' and 'q' are the
n2 probability of success, failure of a game in which
is m 1 A, B and C play in order if A starts the game then,
2
p 1
 A coin is tossed (m+n) times (m>n), then the i ) probability of A’s win = 3 =
1  q 1  q  q2
probability of getting exactly 'm' consecutive heads
n2 qp q
is m  2 ii) probability of B’s win = 3 =
2 1  q 1  q  q2
 When 2 fair dice are rolled, the number of q2 p q 2

favourable cases to get sum ‘k’ is given by iii) probability of C’s win = 3 =
1  q 1  q  q2
 k  1 for 2  k  7 iv) the ratio of their success = 1 : q : q 2

13  k for 8  k  12
 Probability regarding ‘n’ letters and ‘n’
 When 3 fair dice are rolled, the number of envelopes : If 'n' letters are put at random in the
favourable cases to get sum ‘k’ is given by 'n' addressed envelopes, the probability that
1
1 i) All the letters are in right envelopes is
 2  k  1 k  2  for 3  k  8
n!
 ii) Exactly one letter in wrong envelope is 0
 25 for k  9,12 iii) At least one letter may be in wrongly addressed

 27 for k  10,11 1
1 envelope is 1 
n!
  20  k 19  k  for 13  k  18
2 iv) Exactly 'r' ( r  1) letters are in wrong envelopes is
n
pr  1 1 1 (1) r 
 When r fair dice are rolled, the number of favourable  1     ......  
cases to get sum ‘k’ n!  1! 2! 3! r! 
= coefficient of x k in (x1 + x2 + .......+x6)r v) All the letters may be in wrong envelopes
 If r squares are selected at random from a chess pn  1 1 1
n
(1) n 
board the probability that they lies on a diagonal is  1     ......  
n !  1! 2! 3! n! 
line is
4[7Cr  ...........  rCr ]  2(8Cr )  Suppose a word is given. In that let the number of
 (1  r  7)
64Cr vowels be V, number of consonants be C and the
total number of letters be T. If the letters of the
2 word are arranged at random then the probability
= 64 if r  8
C8 that
i) Relative positions of vowels and consonants do not

6
 PROBABILITY

V!C!
change = 3n  3 
n
T! iv) Q is a subset of P is n   
ii) The order of the vowels do not change 4 4
1  If m things are distributed among M persons, then
= no. of ways of arranging the vowels the probability that a particular person get r things
iii) The order of the vowels as well as consonants do
m cr  ( M  1) M r
not chanage (r < m) is
Mm
1

(no. of ways of arranging vowels)(no. of  If n men among whom A and B sit along a circle
ways of arranging consonants) then the odds aganist A and B sitting
iv) The order of the consonants do not change n3 2
together = :  n 3: 2
1 n 1 n 1
= no. of ways of arranging consonants
 A man has a bunch of n keys of which only one fits
 Out of ‘n’ pairs of shoes if r(<n) shoes are selected the door. He tries them successively without
at random then the probability that repetition. Then, the probability that the door is
n
Cr .2 r 1
i) there is no pair  opened at any trail ( less than n)=
2n
Cr n

n
Cr .2r  If ‘n’ men among whom A and B sit along a circle
ii) there is atleast one pair = 1  2n then the probability that there will be exactly ‘r’
Cr
2
Ck .n  k Cr  2 k 2r  2 k
n
persons between A and B = (in both
iii) there are exactly k pairs  2n n 1
Cr
directions)
 Let A be a set containing ‘n’ elements. Then
the probability that ‘r’ subsets of A selected at  Using the vertices of a polygon having n sides a
 2r  1 
n triangle is constructed at random. The probability
random, have no element in common is  r  that the triangle so formed is such that no side of
 2 
 n  4  n  5 
the polygon is side of the triangle is n  1 n  2
 A is a set containing n elements. A subset P of A is   
chosen at random. The set A is reconstructed by  Out of n persons sitting at a round table, three
replacing the elements of P, a subset Q of A is persons are selected at random then the probability
again chosen at random. The probability that that no two of them are consecutive is

i) P  Q contains exactly r elements is

Cr 3n  r n
 n  4  n  5 
4n  n  1 n  2 
3n
ii) P  Q = A is  The probability that two friends will be able to meet
4n
each other between ‘a’ hours and ‘b’ hours, not
n n
2  1  waiting for each other more than ‘c’ hours is
iii) P  Q and P  Q =  is n
  
4  2 
 b  a   1  c 
2 2

b  a 
2

7
 PROBABILITY

RANDOM VARIABLES
 If n whole numbers are taken at random and
multiplied together then the probability that the last  Sample Space (S) : The set of all possible
digit of the product is elementary events in a random trial or experiment
is called sample space for that trial and it is denoted
4n
  2 / 5 by S.
n
i) either 1,3,7 or 9 is n
10  Discrete Sample Space : A sample space S is
called discrete if it is countable or having finite
8n  4 n 4 n  2 n
ii) either 2,4,6 or 8 is  number of sample points.
10n 5n For example:
5n  4 n i) In rolling of a die, S = {1, 2, 3, 4, 5, 6} and it is a
iii) 5 is discrete sample space.
10n
ii) In tossing of a coin, S = {H, T} and it is a discrete
10  8  5  4
n n n n
sample space.
iv) 0 is  Continuous Sample Space : If the number of
10 n
 Let S be a finite set containing ‘n’ elements.Then sample points in a sample space is not countable
then it is called as continuous sample space. For
i) The total number of binary operations on ‘S’ is equal
example:
to n n 2 S = {all possible real values in the interval 1 to 2}
ii) The total number of commutative binary operations  Random Variable : Let S be a sample space
n(n 1) associated with a random experiment. Then a real
on ‘S’ is equal to valued function X : S  R is called a random
n 2
function or random variable.
iii) The probability of selecting a cumulative binary
 Discrete Random Variable : A real valued
n ( n 1) function defined on discrete sample space S is called
n 2 a discrete random variable.
operation from a set S to itself = 2
nn  Continuous Random Variable : A random
variable X defined on continuous sample space S,
iv) The probability of selecting an injective mapping
which can take all real values in an interval (a, b) is
n! called a continuous random variable.
from a set ‘S’ to itself is
nn  Discrete Random Variable : If a discrete
v) The probability of selecting a bijection mapping from variable X can assume values X 1 , X 2 , X 3 ,.... X n with
n! respective probabilities P  X1  , P  X 2  , P  X 3 
a set ‘S’ to itself is
nn ..... P  X n  such that P  X i   0, i and P  X i   1 ,
then X is said to be a discrete random variable.
 Probability Mass Functions : If X is a
discrete random variable which can assume values
X i ; i  1, 2, 3,..... with respective probabilities Pi ;

i  1, 2, 3..... such that  Pi  1, then
i 1

P  X  xi   Pi ; i  1, 2, 3,..... is called probability

mass function of a discrete random variable X. (OR)
If any function P(X=x) gives the probabilities of
various values of a discrete random variable X in
its range, then that function is called probability mass
function.
8
 PROBABILITY
 Probability Distribution : The set of ordered Bernoullian Trials (or) Bernoulli Trials
pairs  xi , P  xi  is called the probability distribution Random trials which result either in the success or
failure of an event A, with constant probability of
of a discrete random variable X.
success p and that of failure 1-p=q are called as
MEAN & VARIANCE bernoullian trials. For example:
 If  xi , P  xi  is the probability distribution of a i) In tossing of an unbiased coin, if we consider getting
discrete random variable X, then its: head upwards as a success then the probability of
Mean or Average ( x or  ) : Expected value of x 1
success p  . The probability of failure
2
or mathematical expectation of x:E(x) or First
moment about origin: 1 '(0) is defined as 1 1
q  1  and it is true for every trial.
2 2
x    1 '(0)  xi P  xi  .
ii) In rolling of a symmetrical die, if we consider getting
2 moment about origin =  2 '(0) is defined as
nd
1
a face with 6 points upward as a success then p 
 2 '(0)  E  X 2
  x P  x  .
2
i i
6

Variance (  2 ) or second moment about mean or 1 5

and q  1   and it is also true for every trial.
6 6
2nd central moment = 2 .
 Binomial Distribution : The probability of x
 2  2  E  x  E  x 
2 successes in n independent bernoullian trials is given
by p  X  x  ncx px qnx ; x  0,1,2,3,.....n ;
E  x 2    E  x 
2

p  0, q  0 and p  q  1 and it is called as B.D.
 2  0   1  0 
1 1 2

Here n & p are called as the parameters of B.D.
 xi P  xi   xi P  xi 
2 2
 A discrete random variable x is said to follow B.D.
with parameters n, p, if its probability mass function
STANDARD DEVIATION
is given by p  X  x   n cx p x qn x ; x  0,1, 2,3,.....n ;
 The positive square root of variance   is called 2

the standard deviation   . p  0, q  0 and p  q  1 . If X is the number of

Cumulative Distribution Functions (CDF) successes in n trials, then the probability distribution
of X is
(or) Distribution Function (DF)
 If X is a discrete random variable, then the No. of
successes r n
0 1
probability that X takes values less than or equal to (x)

a particular value xk (in its range) is called CDF or P(X = x) n

C0qnP0 nC1qn1 P1
n n r
Crq P
r n
Cn q0 Pn
simply DF and it is denoted by F(x). Symbolically
 The above probabilities are various terms of the
k
F  X   P  X  xk    P  X  xi  binomial expansion  q  p n  n c0 q n p 0  n c1q n 1 p1
i 1

k
Here, i1 P  X  xi   1  nc2qn2 p2  .....  ncr qnr pr  .......  ......  ncnqnn pn

 If the mean of the random variable x is x , then the  The originator of B.D. was James Bernoulli (1654-
mean of the random variable aX  b , where a and b 1705) and so it is also some times called as Bernoulli
distribution.
are constants is ax  b .
 If the variance of the random variable x is  2 , then  B  r ; n, p  means p  X  r 
the variance of the random variable aX  b , where The Characteristics of B.D. :
a and b are constants is a 2   2 .  The mean of random binomial variate X is np. i.e.,
 The mathematical expectation of sum of points when x x
x  np or np  x .  n  p , p  n
7n
n symmetrical dice is thrown is
2  The variance of r.b.v. x is npq. i.e.,  2  npq
9
 PROBABILITY

 2  npq 
 Even though n   , p  0 , np    m , a constant
Now q   always. This constant must be a finite positive
x  np 
e   x
 The S.D. of r.b.v. x is npq i.e.,   npq number. Then nlim n
Cx p x qn x  ,   0 and this
 x
 In binomial distribution x   2 or  2  x is called as poisson distribution.
i.e., np  npq or npq  np
e   x
1  The P.D. is P  X  x   x ; x  0,1, 2,3, 4,......
 If p  q  , then the distribution is said to be a
2 Here  is called the parameter of P.D. It was
symmetrical binomial distribution. derived in 1837 by a French mathematician by
MODE : The mode is that value of variable with name S.D. Poisson.
maximum probability.  A discrete random variable X is said to follow
 The mode of B.D. depends on the value of np+p. poisson distribution with parameter  if its
CASE-1: If np+p=k, where k is an integer, then probability mass function is given by
there will be two modes namely k & k-2. In this
e  x
case the distribution is said to be a Bi-modal P  X  x  ; x  0,1, 2,3,......
x
binomial distribution.
CASE-2: If np+p=k+f, where k is an integer and The Characteristics of P.D.
f is a proper fraction then there will be only one  The mean of random poisson variate (r.p.v) X is
mode namely k. i.e., the integral part of np+p will  . i.e., x    np .
be the mode. In this case the distribution is said to  The variance of r.p.v. X is  . i.e.,  2   .
be uni-modal binomial distribution.
 The S.D. of r.p.v. X is  . i.e.,    .
 If we consider n independent bernoullian trials as
one experiment and if we repeat such an experiment  In poisson distribution mean and variance are equal.
N times, then the expected frequency or the i.e., x   2   .
theoretical frequency of x successes is given by  The expected frequency or the theoretical
frequency of x successes when an experiment is
f  X  x  N.p  X  x
repeated N times in P.D. is given by
 N . n cx p x q n  x , x  0,1, 2,3,....n
and this is called as Binomial frequency distribution. f  X  x   N  P ( X  x)
N  e   x
 In a binomial distribution if '  ' is Mean and ' ' is  ; x  0,1, 2,3,......
x
S.D. then 0     (OR) [0,  ) and this is called as poisson frequency distribution.
 In binomial distribution P(X=r) is maximum when Eg.1:-In a poisson distribution
P(X=0)=P(X=1)=k, then the value of k is
r   np  (where [ ] is greatest integer function)
Sol: Given that, P  X  0   P  X  1
In Binomial distribution we know that
e  0 e   1
   1
P  X  x   Cx p q
n x n x
; x  0,1, 2,3,......n 0 1
p  0, q  0 & p  q  1
 k = P(X = 1) = e 1 = 1/e
In this, let
 The number of independent bernoullian trials n is Ex.2:- If X is a poisson variate and
indefinitely large i.e., n   . P(X = 1) = 2P(X = 2) then P(X = 3) is(E-2013)
Sol :Given that, P(X = 1) = 2P(X = 2)
 The probability of success p is infinitely small i.e.,
p0. e  1 e  2 e1.13 1
2   1; P(X = 3) = =
1 2 3 6e

10
 PROBABILITY
Ex.3:- If m is the variance of P.D., then sum of the
terms in odd places is
Sol: P  X  0   P  X  2   P  X  4   .... ,
 0 2 4 
 e      .....  where   m
 0! 2! 4! 

 m0 m2 m 4 
 e m     .....  = e  m coshm
 0! 2! 4! 

Ex.4:- A factory produces lazor blades and 1 in

500 blades is estimated to be defective. The
blades are supplied in packets of 10. In a
consignment of 10,000 packets, using poisson
distribution, approximately the number of
packets which contain no defective blades is
Sol: Let X be the number of defective blades in a lot of
10 blades. The probability that a blade to be
defective = 1/500.
The average number of defective blades in a lot of
1 1
10 = 10    0.02
500 50
Poission parameter is  = 0.02
e  0
P (no defective blades in a lot of 10) =  e0.02
0

0.02 0.0004 0.000008

= 1 1  2  3
 .........

= 1– 0.02 + 0.0002 = 0.9802

Therefore the number of packets in a consignment
of 10000 packets which contain no defective blade
= 10000 × 0.9802 = 9802

11