Probability Density Functions
Jake Blanchard
Spring 2010
Uncertainty Analysis for Engineers 1
Random Variables
We will spend the rest of the semester
dealing with random variables
A random variable is a function defined
on a particular sample space
For example, if we roll two dice there are
36 possible outcomes – this is the sample
space
The sum of the two dice is the random
variable
Uncertainty Analysis for Engineers 2
Random Variables
Let y1 and y2 represent the values of the
two dice
Let x=y1+y2
x can take on any one of 11 values
between 2 and 12, with some more
common than others
The relative likelihood of rolling each of
the possible sums is
2 3 4 5 6 7 8 9 10 11 12
1 2 3 4 5 6 5 4 3 2 1
Uncertainty Analysis for Engineers 3
Probability Distribution Function
We can calculate a probability from this
table and plot the probability against the
sum
0.18
0.16
0.14
Probability of Occurrence
0.12
0.1
0.08
0.06
0.04
0.02
0
2 3 4 5 6 7 8 9 10 11 12
Sum of Two Dice
Uncertainty Analysis for Engineers 4
Continuous Probability Distribution
Functions
Define the pdf [f(x)]such that the
probability that x falls between a and b is
given by
b
Pr(a x b) f ( x)dx
a
Uncertainty Analysis for Engineers 5
Cumulative Probability
What if we are interested in the
probability that the sum is at or below
some value
For example, the probability that the sum
is less than or equal to 4 is
6/36=1/6=0.167
We can plot this value as a function of the
sum
Uncertainty Analysis for Engineers 6
Cumulative Probability
1
0.9
0.8
0.7
Cumulative Probability
0.6
0.5
0.4
0.3
0.2
0.1
0
1 2 3 4 5 6 7 8 9 10 11
Sum of Two Dice
Uncertainty Analysis for Engineers 7
Cumulative Probability
We call this the cumulative distribution
function (CDF)
It has a minimum of 0, a maximum of 1,
and is monotonic
For the example of the sum of two dice,
the CDF is
F xi p( xi )
x xi
x
or F ( x) f ( z )dz
Uncertainty Analysis for Engineers 8
Continuous Functions
Consider the decay of a radioactive
particle
The probability it will survive beyond time
ti is Pr(t>ti)=exp(-ti)
Hence, the CDF is given by
Pr(t<=ti)=F(ti)=1-exp(-ti)
This is plotted for =1/s on the next slide
Uncertainty Analysis for Engineers 9
CDF for radioactive decay
1
0.9
0.8
0.7
0.6
F(time)
0.5
0.4
0.3
0.2
0.1
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
time
Uncertainty Analysis for Engineers 10
Decay Example
For =0.1, the probability that a particle
will decay between 4 and 5 seconds is
given by P(4<t<=5)=F(5)-F(4)=[1-exp(-
0.5)]-[1-exp(-0.4)]=0.063
Uncertainty Analysis for Engineers 11
Characterizing Distributions
Functions
We will see later how to characterize
these functions using
◦ Mean
◦ Median
◦ Standard Deviation
◦ Skewness
◦ Kurtosis
◦ Etc.
Uncertainty Analysis for Engineers 12
Bivariate Distributions
Sometimes we work with more than one
random variable.
These can be correlated, so it is
appropriate to define a single pdf that
governs both variables simultaneously
We call this a joint probability density
function
Uncertainty Analysis for Engineers 13
Joint PDFs
Two continuous random variables are said
to have a bivariate or joint pdf f(x,y) if
y 2 x2
Pr x1 x x2 and y1 y y2 f ( x, y)dxdy
y1 x1
f ( x, y ) 0
f ( x, y)dxdy 1
y1 x1
F x1 , y1 f ( x, y)dxdy
Uncertainty Analysis for Engineers 14
Types of pdfs
We have many choices for functional
forms of pdfs
Our goal is to represent reality
Ultimately, we need data to validate our
choice of pdf
We’ll discuss this later
Next, we’ll look at some of the common
forms
Uncertainty Analysis for Engineers 15
