You are on page 1of 4

Chapter 3 Random Variable

Bernoulli Random Variable


3.1 For a given sample space of some
experiment, a random variable is any
Discrete Random rule that associates a number with Any random variable whose only
Random
Variables and each outcome in . possible values are 0 and 1 is called a

Probability Variables We use X, Y, ... to denote random variables, use


Bernoulli random variable.
x,y,... to represent particular values of a random

Distributions variable.

Types of Random Variables Parameter of a Probability Distribution


3.2 Probability Distribution
A discrete random variable is an rv whose Suppose that p(x) depends on a quantity
that can be assigned any one of a number of
possible values either constitute a finite set Probability Distributions The probability distribution or possible values, each with different value
or else can listed in an infinite sequence. probability mass function (pmf) of a
A random variable is continuous if its set for Discrete Random discrete rv is defined for every number x
determining a different probability
distribution. Such a quantity is called a
of possible values consists of an entire
interval on a number line.
Variables by p(x) = P (all s ∈ :X ( s ) = x). parameter of the distribution. The
collection of all distributions for all different
parameters is called a family of
distributions.

Cumulative Distribution Function Proposition Probability Distribution for the


Random Variable X
The cumulative distribution function (cdf) 3.3
For any two numbers a and b with a ≤ b, A probability distribution for a random variable X:
F(x) of a discrete rv variable X with pmf p
P(a ≤ X ≤ b) = F (b) − F (a −)
(x) is defined for every number by x –8
P(X = x) 0.13
–3
0.15
–1
0.17
0
0.20
1
0.15
4 6
0.11 0.09
Expected Values of
“a–” represents the largest possible X
F ( x) = P( X ≤ x) =
y: y ≤ x
p( y)
value that is strictly less than a. Discrete Random
Find
For any number x, F(x) is the probability Note: For integers a. P ( X ≤ 0 ) 0.65 Variables
that the observed value of X will be at b. P ( −3 ≤ X ≤ 1)
P(a ≤ X ≤ b) = F (b) − F (a − 1) 0.67
most x.

Ex. Use the data below to find out the expected The Expected Value of a Function
The Expected Value of X number of the number of credit cards that a student
Rules of the Expected Value
will possess.
Let X be a discrete rv with set of If the rv X has the set of possible values E (aX + b) = a ⋅ E ( X ) + b
x = # credit cards
possible values D and pmf p(x). The D and pmf p(x), then the expected
expected value or mean value of X,
x P(x =X) E ( X ) = x1 p1 + x2 p2 + ... + xn pn value of any function h(x), denoted This leads to the following:
0 0.08
denoted E ( X ) or µ X , is 1 0.28
= 0(.08) + 1(.28) + 2(.38) + 3(.16) E[h( X )] or µh ( X ) , is 2. For any constant a,
+ 4(.06) + 5(.03) + 6(.01)
E ( X ) = µX = x ⋅ p (x )
2 0.38
E[h ( X )] = h (x) ⋅ p (x) E (aX ) = a ⋅ E ( X ).
3 0.16
x∈D
=1.97 D
4 0.06
5 0.03 About 2 credit cards
2. For any constant b,
6 0.01 E ( X + b) = E ( X ) + b.
The Variance and Standard Ex. The quiz scores for a particular student are given 2 2
V ( X ) = .08 (12 − 21) + .15 (18 − 21) + .31( 20 − 21)
2
below: Shortcut Formula for Variance
Deviation 22, 25, 20, 18, 12, 20, 24, 20, 20, 25, 24, 25, 18
2 2
+.08 ( 22 − 21) + .15 ( 24 − 21) + .23( 25− 21)
2

Let X have pmf p(x), and expected value µ Find the variance and standard deviation.
V ( X ) = 13.25
Then the variance of X, denoted V(X) Value 12 18 20 22 24 25 V (X ) = σ 2 = x2 ⋅ p ( x ) − µ 2
(or σ X2 or σ 2 ), is Frequency 1 2 4 1 2 3 σ = V (X ) = 13.25 ≈ 3.64 D

V (X ) = ( x − µ ) 2 ⋅ p ( x ) = E [( X − µ ) 2 ]
Probability
µ = 21
.08 .15 .31 .08 .15 .23
( )
= E X2 − E(X )
2

D 2 2 2
V ( X ) = p1 ( x1 − µ ) + p2 ( x2 − µ ) + ... + pn ( xn − µ )
The standard deviation (SD) of X is
σ = V (X )
σ X = σ X2

Rules of Variance Binomial Experiment


An experiment for which the following 1. The trials are identical, and each trial
2 2 2
V (aX + b) = σ aX +b = a ⋅ σ X 3.4 four conditions are satisfied is called a can result in one of the same two
and σ aX +b = a ⋅ σ X binomial experiment. possible outcomes, which are denoted
The Binomial 1. The experiment consists of a
by success (S) or failure (F).
This leads to the following:
2
Probability sequence of n trials, where n is fixed in
2. The trials are independent.
1. σ aX = a 2 ⋅ σ X2 , σ aX = a ⋅σ X advance of the experiment. 3. The probability of success is constant
2. σ X +b = σ X2
2 Distribution from trial to trial: denoted by p.

Binomial Experiment Binomial Random Variable Computation of a


Notation for the pmf
Suppose each trial of an experiment can of a Binomial rv Binomial pmf
Given a binomial experiment consisting
result in S or F, but the sampling is of n trials, the binomial random variable
without replacement from a population of X associated with this experiment is n n− x
Because the pmf of a binomial rv X p x (1 − p ) x = 0,1, 2,...n
size N. If the sample size n is at most 5% defined as b ( x; n, p ) = p
of the population size, the experiment can depends on the two parameters n and
p, we denote the pmf by b(x;n,p). 0 otherwise
be analyzed as though it were exactly a X = the number of S’s among n trials
binomial experiment.

Ex. A card is drawn from a standard 52-card deck. Ex. 5 cards are drawn, with replacement, from a
If drawing a club is considered a success, find the Notation for cdf standard 52-card deck. If drawing a club is
probability of Mean and Variance considered a success, find the mean, variance, and
a. exactly one success in 4 draws (with replacement). standard deviation of X (where X is the number of
For X ~ Bin(n, p), the cdf will be successes).
p = ¼; q = 1– ¼ = ¾
For X ~ Bin(n, p), then E(X) = np, V p = ¼; q = 1– ¼ = ¾
denoted by
4 1
1
3
3
x (X) = np(1 – p) = npq, σ X = npq
1
1 4 4 ≈ 0.422 P( X ≤ x ) = B ( x; n, p ) = b( y ; n , p ) (where q = 1 – p). µ = np = 5 =1.25
4
b. no successes in 5 draws (with replacement). y =0
1 3
V ( X ) = npq = 5 = 0.9375
5 1
0
3
5 x = 0, 1, 2, …n 4 4
≈ 0.237
0 4 4 σ X = npq = 0.9375 ≈ 0.968
Ex. If the probability of a student successfully passing
this course (C or better) is 0.82, find the probability
The Hypergeometric Distribution
that given 8 students 1. Each individual can be
3.5 The three assumptions that lead to a characterized as a success (S) or
8
a. all 8 pass. ( 0.82 ) ( 0.18) ≈ 0.2044
8 0
hypergeometric distribution: failure (F), and there are M
8
successes in the population.
b. none pass. 8
( 0.82) 0 ( 0.18) 8 ≈ 0.0000011 Hypergeometric and 1. The population or set to be sampled
0 consists of N individuals, objects, or 2. A sample of n individuals is
c. at least 6 pass. Negative Binomial elements (a finite population). selected without replacement in
8
( 0.82 )6 ( 0.18) 2 +
8
( 0.82 )7 ( 0.18 )1 +
8
( 0.82 )8 ( 0.18) 0 Distributions such a way that each subset of size
6 7 8 n is equally likely to be chosen.
≈ 0.2758 + 0.3590 + 0.2044 = 0.8392

Hypergeometric Distribution The Negative Binomial Distribution


If X is the number of S’s in a completely Hypergeometric Mean and
random sample of size n drawn from a Variance The negative binomial rv and distribution 3. The probability of success is constant
population consisting of M S’s and (N – M) are based on an experiment satisfying the from trial to trial, so P(S on trial i) = p for
F’s, then the probability distribution of X, following four conditions: i = 1, 2, 3, …
called the hypergeometric distribution, is M N −n M M
E( X ) = n⋅ V (X ) = ⋅ n⋅ 1− 1. The experiment consists of a sequence
given by M N −M N N −1 N N 4. The experiment continues until a total
of independent trials. of r successes have been observed, where
x n− x
P( X = x ) = h ( x; n, M , N ) = 2. Each trial can result in a success (S) or r is a specified positive integer.
N
n a failure (F).
max(0, n − N + M ) ≤ x ≤ min(n, M )

Poisson Distribution
pmf of a Negative Binomial
Negative Binomial
Mean and Variance
3.6
The pmf of the negative binomial rv X A random variable X is said to have
with parameters r = number of S’s and a Poisson distribution with
p = P(S) is r (1 − p) r (1 − p) The Poisson Probability parameter λ ( λ > 0 ) , if the pmf of X
E(X ) = V (X ) =
x + r +1 p p
2
Distribution is
nb ( x; r , p) = p r (1 − p) x e −λ λ x
r −1 p ( x; λ ) = x = 0,1, 2...
x!
x = 0, 1, 2, …

The Poisson Distribution Poisson Process


Poisson Distribution
as a Limit
Mean and Variance 1. The probability of more than one
3 Assumptions: event during ∆t is o ( ∆t ) .
Suppose that in the binomial pmf b(x;n, p),
If X has a Poisson distribution with 1. There exists a parameter α > 0 such
we let n → ∞ and p → 0 2. The number of events during the time
parameter λ , then that for any short time interval of length
interval ∆t is independent of the
in such a way that np approaches a value ∆t, the probability that exactly one event
E( X ) = V ( X ) = λ number that occurred prior to this time
λ > 0. is received is α ⋅ ∆t + o ( ∆t ) .
interval.
Then b( x; n, p ) → p ( x; λ ).
Poisson Distribution
Pk (t ) = e−α t ⋅ (α t ) k / k !, so that the number
of pulses (events) during a time interval of
length t is a Poisson rv with parameter
λ = α t. The expected number of pulses
(events) during any such time interval is
α t , so the expected number during a unit
time interval is α .

You might also like