Random Variables and

Probability Distribution

Purnomo
Jurusan Teknik Mesin
UGM

1
Random Variables
A random variable X is a numerical valued
function defined on a sample space.
A number X(e), providing a measure of
characteristic of interest, is assigned to
each simple event e in the sample
space
Contoh dadu :
X = 1, 2, 3, 4, 5, 6

2
 Two balls are drawn in succession from a box that contains 4
red balls and 3 blue balls. The possible outcomes and the values
y of the random variable Y, where Y is the number of red balls
is

Sample space y
RR 2
RB 1
BR 1
BB 0

3
Illustration
Two products A and B are judge by four consumer who then
expressed a preference for A and B. The outcome when the first
and third consumers prefer A and the other consumers prefer B
is denoted by ABAB. The number of outcomes is 24 = 16.

AAAA AAAB AABB ABBB BBBB

AABA ABAB BABB
ABAA ABBA BBAB
BAAA BAAB BBBA
BBAA
BABA

4
Illustration
Suppose that the products are alike in quality and that the
consumers express their preference independently. Then the 16
simple events in the sample space are equally likely, and each
has a probability of 1/16. Let a random variable X be devined as
X= number of person preffering A to B.
Probability distribution :

Distinct value of X 0 1 2 3 4
Probability 1/16 4/16 6/16 4/16 1/16

P[X≥2] = 6/16 + 4/16 + 1/16 = 11/16

P[1≤X≤3] = 4/16 + 6/16 + 4/16 = 14/16

5
Probability Distribution

The probability distribution or simply, the distribution of

a discrete random variable is a list of the distinct
values of xi together with their associate probabilities
f(xi) = P[X=xi]

6
Graphic Presentation
f(x)
Line diagram
6/16

4/16

2/16

x
0 1 2 3 4 7
Histogram of probability
 Histogram
Value x 1 2 3 4
f(x) 1/8 ¼ ½ 1/8

4/8
Area = 0.5

2/8

8
1 2 3 4 5
Properties of relative frequency histogram

 The total area under the histogram is 1

 For the two points a and b such that each is a
boundary point of some class, the relative
frequency of measurements in the interval a
to b is the area under the histogram enclosed
by this interval.

9
Expectation
Expected value or Expectation of X

E ( X )   xi f ( xi )

X 0 1 2 3 4 5 Total
f(x) 0.1 0.1 0.2 0.3 0.2 0.1 1
xf(X) 0 0.1 0.4 0.9 0.8 0.5 2.7

E(X) = 2.7

E(X) = population mean = μ

10
Variance : a measure of spread
Deviation = X – μ
(x1 – μ), (x2 – μ), …………………………….(xk – μ)
Probabilities
f(x1), f(x2), ………………………………….f(xk)

E(deviation) = E(X – μ) = ∑(xi – μ)f(xi) = 0

Deviation can not be used as a measure of spread

11
Variance and standard
Deviation

Variance of X (= σ2 = σx2)
Var(X) = E[(X – μ)2] = E(X2) – μ2

Standard deviation ( = σ = σx )

sd(X) = √ Var (X)

12
Standardized random variable

Standardized random variable :

X  k
Z has E(Z) = 0 and Var(Z) = 1
k
Random variable Z has a mean of 0 and
variance of 1
Bentuk ini akan banyak digunakan pada applikasi 13
PROBABILITY MODELS FOR CONTINUOUS
RANDOM VARIABLES

 The probability distribution of a

continuous random variable can be
visualized as a smooth form of relative
frequency histogram based on large
number of observations.

14
PROBABILITY DENSITY CURVE

 Probability density curve can be viewed

as a limiting form of relative frequency
histogram
(number of classes - infinite )

15
Properties of Probability Density Function, f(x)

 The total area under the density curve is 1

P a  X  b  
 area under the
density curve between a and b
 f(x) is positive or zero
 For continuous random variable, the
probability that X=x is always 0 (X is only
meaningful when X lies in an interval

16
Density Curves
 Measuring center and spread for density curves
 Density curves describe the overall shape of a
distribution
 Ideal patterns that are accurate enough for
practical purposes
 Faster to draw and easier to use
 Areas or proportions under the curve represent
counts or percents of observations

17
Features of a Continuous Distribution

 As with relative frequency histograms, the probability

density curves of continuous random variables
posses a wide variety of shapes :
- Negatively skewed
- Symmetric
- Positively skewed
- Flat
- Bell shaped
- Peaked

18
Center of a Density Curve
 The mode of a distribution is the point where
the curve is highest
 The median is the point where half of the
area under the curve lies on the left and the
other half on the right. Equal Areas Point
 Quartiles can be found by dividing the area
under the curve into four equal parts
 ¼ of the area is to the left of the 1st quartile

3 of the area is to the left of the 3rd quartile
4
 The mean is the balance point.
19
Percentiles

 Percentiles are defined as :

The population 100p-th percentile is an x value that has
an area p to the left and 1-p to the right.

Lower (first) quartile = 25th percentile

Second quartile (or median) = 50th percentile
Upper (third) quartile=75th percentile

20
The Normal distribution
 Discovered by Abraham de Moivre around 1720. Around 1870,
Adolph Quetelet realised that the normal curve could be used to
compare histograms of data.
 Chest measurements of 5738 Scottish soldiers by Belgian
scholar Lambert Quetelet (1796-1874)
 Pierre Laplace dan Carl Gauss : bell-shaped distribution
Gauss derived the normal distribution mathematically as the
probability distribution of the error of measurements, which is
called “normal law of error”
 Gaussian Distribution

21
Normal Distributions
 Symmetric
 Single-peaked (unimodal)
 Bell-shaped
 The mean, median, and mode are the same
 The points where there is a change in
curvature is one standard deviation on either
side of the mean.
 The mean and standard deviation completely
specify the curve
22
Normal Distribution

 The height of a normal

density curve at any point x
1  2
 12 ( x )
is given by f ( x)  e
 2
 is the mean
 is the standard deviation

 N ( , )

 23
Normal Distribution
 The probability of the
interval extending

P     X       0.683
P   2  X    2   0.954
P   3  X    3   0.997

24
The Empirical Rule
 68% of the observations fall within one
standard deviation of the mean
 95% of the observations fall within two
standard deviation of the mean
 99.7% of the observations fall within
three standard deviation of the mean

25
Example: Young Women’s
Height
 The heights of young women are approximately
normal with mean = 64.5 inches and std.dev. = 2.5
inches.

26
The normal distribution is the most important distribution
in Statistics. Typical normal curves with different sigma
(standard deviation) values are shown below.

27
Examples with approximate
Normal distributions
 Height
 Weight
 IQ scores
 Standardized test scores
 Body temperature
 Repeated measurement of same quantity
 …
28
FACTS
 Universality of the normal distribution is
only a myth, and examples of quite
nonnormal distribution abound in any
virtually every field of study
 Still, the normal distribution plays a
central role in statistics (make things
easier)

29
Standardizing and z-Scores
 One case, one curve --- too
complicated
 Solution -- standardization

normalization
non-dimensionalization
---- z-Scores
---- All cases, one curve (or table)
30
Standardizing and z-Scores
 an observation x comes from a distribution with
mean µ and standard deviation σ
 The standardized value of x is defined as

x
z  ,

which is also called a z-score.
 A z-score indicates how many standard deviations
the original observation is away from the mean,
and in which direction.

31
The Standard Normal Curve
N(0,1)

32
Standardizing and z-Scores

N   , 
X 
Z ,

X  b  b
P X  b   P     P Z 
      
a   X   b    a   b
P a  X  b   P      P Z
        

Za Zb
33
The Standard Normal Table
 The Normal Table is a table of areas under the
standard normal density curve. The table entry for each
value z is the area under the curve to the left of z.

34
The Standard Normal Table
 The Normal Table can be used to find the proportion of
observations of a variable which fall to the left of a specific
value z if the variable follows a normal distribution.

35
36
37
38
Use of The Normal Table

P Z  z   Area under curve to the left of z

P a  z  b   (area to the left of b)- (area to the

left of a)

39
Use of The Normal Table
 Symmetry
P Z  0  0.5

P Z   z   1  P Z  z 

40
Use of The Normal Table
 If z>0

P Z  z   0.5  P 0  Z  z 

P Z   z   0.5  P 0  Z  z 

41
Use of The Normal Table
 Calculate z
 Find the area to the left of z in Standard
Normal Probability Table
 Other calculations obey the properties
of the Standardized Normal Curve

42
Example : random variable

43
Example : Expectation

44
45
46
47
48