PROBABILITY 
DISTRIBUTION
Random Variable 
 Way to map outcomes of random process to numbers. 
 Random process is a phenomenon that varies to some degree 
unpredictably as time goes on. If we observed an entire time-sequence 
of the process on several different occasions, under 
presumably “identical” conditions, the resulting observation 
sequences, in general, would be different. 
 Examples are flipping a coin, rolling a die. 
푋 = 
1 푖푓 ℎ푒푎푑 푓푎푙푙 
0 푖푓 푡푎푖푙 푓푎푙푙 
 Mapped the outcome of random process (flipping 
coin) and quantified to some variable. (Random 
variable)
Types of Random Variable 
 Discrete Random Variable 
 Distinct and separate value. (Can list all the possible value. 
i.e., countable) 
 푋 = 
1 푖푓 ℎ푒푎푑 푓푎푙푙 
0 푖푓 푡푎푖푙 푓푎푙푙 
(possible values are 1 and 0) 
 Continuous Random Variable 
 Any value in an interval. Interval may be infinite also. 
(Uncountable) 
 Ex: Mass of random animals selected at Zoo. (Possible 
outcomes are uncountable)
Discrete Probability Distribution 
 X = number of “heads” after 3 flips of fair coins 
 X is Discrete random variable ( outcomes are 0,1,2 
and 3) 
 Fair coins outcomes : TTT, TTH, THT, THH, HTT, 
HTH, HHT, HHH. 
 P(X=0) = 1/8 
 P(X=1) = 3/8 
 P(X=2) = 3/8 
 P(X=3) = 1/8 
 This is called as Discrete probability distribution.
Continuous Probability 
Distribution 
 In continuous case, the function f(x) is called 
the probability density function, and 
probabilities are determined by the areas 
under the curve f(x). 
 Ex: X= Exact amount of rain tomorrow. (may be 1 
inches, 2,2.001, 3.1 etc.) 
 Imagine that probability for all values has given.
Continuous Probability 
Distribution 
 If want to know, what is the probability exactly 2 
inches of rain? P(X=2) not 2.0001,1.9999 
,2.0000001.not possible. 
 So we can say 푃( 푋 − 2 < 표. 1). It means 
P(1.9<X<2.1). It’s nothing but some area in the 
graph. 
 Based on integral rule, so probabilities are 
determined by the areas under the curve.
Discrete Probability Distribution vs 
Probability Density Function (PDF) 
 In Discrete, distribution is Discrete Probability 
Distribution and in continuous, the distribution 
is PDF. 
 In PDF, we find probability for some interval 
only. (Area of the line is zero) 
 Sum of all the events probability for both 
discrete and continuous is one.
Expected value 
 The mean (or expected value) of X (is random 
variable) gives the value that we would expect 
to observe on average in a large number of 
repetitions of the experiment 
X i x X P x X E     
( ) * ( ) 
1 
i 
n 
i 

Example 
 An investment in Project A will result in a 
loss of $26,000 with probability 0.30, break 
even with probability 0.50, or result in a profit 
of $68,000 with probability 0.20. An 
investment in Project B will result in a loss of 
$71,000 with probability 0.20, break even 
with probability 0.65, or result in a profit of 
$143,000 with probability 0.15. Which 
investment is better?
Solution 
 Random Variable (X)- The amount of money received from 
the investment in Project B X can assume only x1 , x2 , x3 
 X= x1 is the event that we have Loss 
 X= x2 is the event that we are breaking even 
 X= x3 is the event that we have a Profit 
 x1=$-71,000 
 x2=$0 
 x3=$143,000 
 P(X= x1)=0.2 
 P(X= x2)= 0.65 
 P(X= x3)= 0.15
Solution 
 The amount of money received from the investment 
in Project B 
 X can assume only x1 , x2 , x3 
 X= x1 is the event that we have Loss 
 X= x2 is the event that we are breaking even 
 X= x3 is the event that we have a Profit 
 x1=$-71,000 
 x2=$0 
 x3=$143,000 
 P(X= x1)=0.2 
 P(X= x2)= 0.65 
 P(X= x3)= 0.15 
Project B is Better. We made decision using Expected value.
Binomial Distribution 
 Imagine a simple trial with only two possible outcomes 
 Success (S) 
 Failure (F) 
 Examples 
 Toss of a coin (heads or tails) 
 Sex of a newborn (male or female) 
 Survival of an organism in a region (live or die)
Binomial Distribution 
 Suppose that the probability of success is p 
 What is the probability of failure? 
q = 1 – p 
Examples 
 Toss of a coin (S = head): p = 0.5 , q = 0.5 
 Roll of a die (S = 1): p = 0.1667 , q = 0.8333 
 Fertility of a chicken egg (S = fertile): p = 0.8 , q = 0.2
Binomial Distribution 
 Imagine that a trial is repeated n times 
Examples: 
 A coin is tossed 5 times 
 A die is rolled 25 times 
 50 chicken eggs are examined 
 Assume p remains constant from trial to trial and that the 
trials are independent of each other
Binomial Distribution 
 What is the probability of obtaining x successes in 
n trials? 
Example: 
 What is the probability of obtaining 2 heads from a 
coin that was tossed 5 times? 
P(HHTTT) = (1/2)^5 = 1/32
Binomial Distribution 
There are other possibilities also 
HHTTT, HTHTT ,HTTHT, HTTTH 
THHTT ,THTHT ,THTTH 
TTHHT ,TTHTH 
TTTHH 
Then the probability of x successes is 
P(x) = q  q  p  q   
= px  qn – x
Binomial Distribution 
 Probability of getting two Head is 
P(HHTTT) = 
ퟏ 
ퟐ 
ퟏ 
ퟐ 
1 
2 
1 
2 
1 
2 
1 
2 
= 
2 1 
2 
3 
 How to include other possibilities like HTHTT 
etc., 
 Order is not important. So multiply with 
combination. n! 
P(x) = px  qn – x 
x!(n – x)! 
Combination gives 
the number of 
ways to obtain x 
successes in n
Binomial Distribution- Overviews 
 It is the discrete probability distribution of the number of 
successes in a sequence of n independent yes/no 
experiments. 
 A success/failure experiment is also called a Bernoulli 
experiment or Bernoulli trial; when n = 1. 
 The binomial distribution is frequently used to model the 
number of successes in a sample of size n drawn with 
replacement from a population of size N. 
 If the sampling is carried out without replacement, the 
draws are not independent and so the resulting distribution 
is a hypergeometric distribution
Example 
 Suppose you independently throw a 
dart 10 times. Each time you throw a dart, the 
probability of hitting the target is 3/4 . What is 
the probability of hitting the target less 
than 5 times (out of the total times you throw a 
dart)?
Solution
Expected value for Binomial 
Distribution 
E(X) = nP 
(Final answer from derivation of expected value 
for Binomial Distribution) 
The Expected times through the dart is 
E(X) = 10*3/4 
= 7.5
Poisson Distribution 
 Discovered by Mathematician Simeon Poisson 
in France in 1781. 
 The modelling distribution that takes his name 
was originally derived as an approximation to 
the binomial distribution. 
 n is very large (trial) (n tends to infinity) 
 P is very small (Probability of success)
Poisson Distribtution 
 It is an eg of a probability model which is usually 
defined by the mean no. of occurrences in a time 
interval and simply denoted by λ. 
 Ex: 
 Number of telephone calls in a week. 
 Number of people arriving at a 
checkout in a day. 
 Number of industrial accidents per 
month in a manufacturing plant.
When to use? 
 Occurrences are independent. 
 Occurrences are random. 
 The probability of an occurrence is constant 
over time.
Poisson Distribution 
 If we substitute 휆 /n for p, and let n tend to 
infinity, the binomial distribution becomes the 
Poisson distribution: 
P(x) = 
e -휆휆x 
x!
Example 
 A production line produces 600 parts per hour 
with an average of 5 defective parts an hour. If 
you test every part that comes off the line in 15 
minutes, what is the probability of finding no 
defective parts? 
 Time interval = 15 minutes 
 Occurrence rate = 5 defects per hour
Solution 
λ = (5 defects/hour)*(0.25 hour) 
= 1.25 
p(x) = (xe-)/(x!) 
x = given number of defects 
P(x = 0) = (1.25)0e-1.25)/(0!) 
= e-1.25 = 0.287 
= 28.7%
Questions 
 Vehicles pass through a junction on a busy 
road at an average rate of 300 per hour. 
 Find the probability that none passes in a given 
minute. 
Solution: 
 The average number of cars per minute is: μ= 300/60=5 cars/min 
P(X) = 0.12511
Normal Distribution 
 The distribution of data happens to be 
perfectly symmetrical. 
 It is perfectly bell shaped curve in which case 
the value of mean 푋 = median M = mode Z.
Mean and Standard Deviation 
 The mean is found by adding all the values in the 
set, then dividing the sum by the number of 
values. 
 It is most widely used measure of dispersion of a 
series and is commonly denoted by the symbol ‘ 
휎’
Mean and Standard Deviation 
Mean 
Standard 
Deviation
Probability distribution for Dummies

Probability distribution for Dummies

  • 1.
  • 2.
    Random Variable Way to map outcomes of random process to numbers.  Random process is a phenomenon that varies to some degree unpredictably as time goes on. If we observed an entire time-sequence of the process on several different occasions, under presumably “identical” conditions, the resulting observation sequences, in general, would be different.  Examples are flipping a coin, rolling a die. 푋 = 1 푖푓 ℎ푒푎푑 푓푎푙푙 0 푖푓 푡푎푖푙 푓푎푙푙  Mapped the outcome of random process (flipping coin) and quantified to some variable. (Random variable)
  • 3.
    Types of RandomVariable  Discrete Random Variable  Distinct and separate value. (Can list all the possible value. i.e., countable)  푋 = 1 푖푓 ℎ푒푎푑 푓푎푙푙 0 푖푓 푡푎푖푙 푓푎푙푙 (possible values are 1 and 0)  Continuous Random Variable  Any value in an interval. Interval may be infinite also. (Uncountable)  Ex: Mass of random animals selected at Zoo. (Possible outcomes are uncountable)
  • 4.
    Discrete Probability Distribution  X = number of “heads” after 3 flips of fair coins  X is Discrete random variable ( outcomes are 0,1,2 and 3)  Fair coins outcomes : TTT, TTH, THT, THH, HTT, HTH, HHT, HHH.  P(X=0) = 1/8  P(X=1) = 3/8  P(X=2) = 3/8  P(X=3) = 1/8  This is called as Discrete probability distribution.
  • 5.
    Continuous Probability Distribution  In continuous case, the function f(x) is called the probability density function, and probabilities are determined by the areas under the curve f(x).  Ex: X= Exact amount of rain tomorrow. (may be 1 inches, 2,2.001, 3.1 etc.)  Imagine that probability for all values has given.
  • 6.
    Continuous Probability Distribution  If want to know, what is the probability exactly 2 inches of rain? P(X=2) not 2.0001,1.9999 ,2.0000001.not possible.  So we can say 푃( 푋 − 2 < 표. 1). It means P(1.9<X<2.1). It’s nothing but some area in the graph.  Based on integral rule, so probabilities are determined by the areas under the curve.
  • 7.
    Discrete Probability Distributionvs Probability Density Function (PDF)  In Discrete, distribution is Discrete Probability Distribution and in continuous, the distribution is PDF.  In PDF, we find probability for some interval only. (Area of the line is zero)  Sum of all the events probability for both discrete and continuous is one.
  • 8.
    Expected value The mean (or expected value) of X (is random variable) gives the value that we would expect to observe on average in a large number of repetitions of the experiment X i x X P x X E     ( ) * ( ) 1 i n i 
  • 9.
    Example  Aninvestment in Project A will result in a loss of $26,000 with probability 0.30, break even with probability 0.50, or result in a profit of $68,000 with probability 0.20. An investment in Project B will result in a loss of $71,000 with probability 0.20, break even with probability 0.65, or result in a profit of $143,000 with probability 0.15. Which investment is better?
  • 10.
    Solution  RandomVariable (X)- The amount of money received from the investment in Project B X can assume only x1 , x2 , x3  X= x1 is the event that we have Loss  X= x2 is the event that we are breaking even  X= x3 is the event that we have a Profit  x1=$-71,000  x2=$0  x3=$143,000  P(X= x1)=0.2  P(X= x2)= 0.65  P(X= x3)= 0.15
  • 11.
    Solution  Theamount of money received from the investment in Project B  X can assume only x1 , x2 , x3  X= x1 is the event that we have Loss  X= x2 is the event that we are breaking even  X= x3 is the event that we have a Profit  x1=$-71,000  x2=$0  x3=$143,000  P(X= x1)=0.2  P(X= x2)= 0.65  P(X= x3)= 0.15 Project B is Better. We made decision using Expected value.
  • 12.
    Binomial Distribution Imagine a simple trial with only two possible outcomes  Success (S)  Failure (F)  Examples  Toss of a coin (heads or tails)  Sex of a newborn (male or female)  Survival of an organism in a region (live or die)
  • 13.
    Binomial Distribution Suppose that the probability of success is p  What is the probability of failure? q = 1 – p Examples  Toss of a coin (S = head): p = 0.5 , q = 0.5  Roll of a die (S = 1): p = 0.1667 , q = 0.8333  Fertility of a chicken egg (S = fertile): p = 0.8 , q = 0.2
  • 14.
    Binomial Distribution Imagine that a trial is repeated n times Examples:  A coin is tossed 5 times  A die is rolled 25 times  50 chicken eggs are examined  Assume p remains constant from trial to trial and that the trials are independent of each other
  • 15.
    Binomial Distribution What is the probability of obtaining x successes in n trials? Example:  What is the probability of obtaining 2 heads from a coin that was tossed 5 times? P(HHTTT) = (1/2)^5 = 1/32
  • 16.
    Binomial Distribution Thereare other possibilities also HHTTT, HTHTT ,HTTHT, HTTTH THHTT ,THTHT ,THTTH TTHHT ,TTHTH TTTHH Then the probability of x successes is P(x) = q  q  p  q   = px  qn – x
  • 17.
    Binomial Distribution Probability of getting two Head is P(HHTTT) = ퟏ ퟐ ퟏ ퟐ 1 2 1 2 1 2 1 2 = 2 1 2 3  How to include other possibilities like HTHTT etc.,  Order is not important. So multiply with combination. n! P(x) = px  qn – x x!(n – x)! Combination gives the number of ways to obtain x successes in n
  • 18.
    Binomial Distribution- Overviews  It is the discrete probability distribution of the number of successes in a sequence of n independent yes/no experiments.  A success/failure experiment is also called a Bernoulli experiment or Bernoulli trial; when n = 1.  The binomial distribution is frequently used to model the number of successes in a sample of size n drawn with replacement from a population of size N.  If the sampling is carried out without replacement, the draws are not independent and so the resulting distribution is a hypergeometric distribution
  • 19.
    Example  Supposeyou independently throw a dart 10 times. Each time you throw a dart, the probability of hitting the target is 3/4 . What is the probability of hitting the target less than 5 times (out of the total times you throw a dart)?
  • 20.
  • 21.
    Expected value forBinomial Distribution E(X) = nP (Final answer from derivation of expected value for Binomial Distribution) The Expected times through the dart is E(X) = 10*3/4 = 7.5
  • 22.
    Poisson Distribution Discovered by Mathematician Simeon Poisson in France in 1781.  The modelling distribution that takes his name was originally derived as an approximation to the binomial distribution.  n is very large (trial) (n tends to infinity)  P is very small (Probability of success)
  • 23.
    Poisson Distribtution It is an eg of a probability model which is usually defined by the mean no. of occurrences in a time interval and simply denoted by λ.  Ex:  Number of telephone calls in a week.  Number of people arriving at a checkout in a day.  Number of industrial accidents per month in a manufacturing plant.
  • 24.
    When to use?  Occurrences are independent.  Occurrences are random.  The probability of an occurrence is constant over time.
  • 25.
    Poisson Distribution If we substitute 휆 /n for p, and let n tend to infinity, the binomial distribution becomes the Poisson distribution: P(x) = e -휆휆x x!
  • 26.
    Example  Aproduction line produces 600 parts per hour with an average of 5 defective parts an hour. If you test every part that comes off the line in 15 minutes, what is the probability of finding no defective parts?  Time interval = 15 minutes  Occurrence rate = 5 defects per hour
  • 27.
    Solution λ =(5 defects/hour)*(0.25 hour) = 1.25 p(x) = (xe-)/(x!) x = given number of defects P(x = 0) = (1.25)0e-1.25)/(0!) = e-1.25 = 0.287 = 28.7%
  • 28.
    Questions  Vehiclespass through a junction on a busy road at an average rate of 300 per hour.  Find the probability that none passes in a given minute. Solution:  The average number of cars per minute is: μ= 300/60=5 cars/min P(X) = 0.12511
  • 29.
    Normal Distribution The distribution of data happens to be perfectly symmetrical.  It is perfectly bell shaped curve in which case the value of mean 푋 = median M = mode Z.
  • 30.
    Mean and StandardDeviation  The mean is found by adding all the values in the set, then dividing the sum by the number of values.  It is most widely used measure of dispersion of a series and is commonly denoted by the symbol ‘ 휎’
  • 31.
    Mean and StandardDeviation Mean Standard Deviation