Probability and Stochastic Processes

Chapter 3 Continuous Random Variables

0-0
 Continuous Random Variables
• Example: Spin wheel with circumference c = 1. Measure distance
around circumference

1
 CDF for continuous RVs
• Definition 3.1: The cumulative distribution function (CDF) of random
variable X is
FX (x) = P [X ≤ x]

• Definition 3.2: X is a continuous random variable if the CDF FX (x) is

a continuous function.

2
 CDF of the Wheel Pointer
• For the Wheel pointer:

X X
Y=n Y=1
Y=2

Y=3

⎧
⎪
⎨ 0
⎪ x<0
FX (x) = x 0≤x≤1
⎪
⎪
⎩
1 x>1

3
 CDF Properties
For any rv X,
• FX (−∞) = 0 and FX (∞) = 1

• P [x1 < X ≤ x2 ] = FX (x2 ) − FX (x1 )

4
 Probability Density Function (PDF)
• Definition 3.3: The probability density function (PDF) of a continuous
random variable X is

fX (x) = dFX (x) /dx

(if FX (x) is differentiable).

• Example 3.3 (MATLAB Demo)

5
 PDF Properties
For a continuous random variable X with PDF fX (x),
• fX (x) ≥ 0 for all x
x
• FX (x) = −∞ fX (u) du
∞
• −∞ fX (x) dx = 1
 x2
• P [x1 < X ≤ x2 ] = x1 fX (x) dx

6
 Expected Values
• Definition 3.4: The expected value of a continuous random variable X is
 ∞
E[X] = xfX (x) dx
−∞

7
 Expected Value of a Function
• The expected value of a function, g(X), of random variable X is
 ∞
E[g(X)] = g(x)fX (x) dx
−∞

8
 Properties of Expected Values
For any random variable X,
• E[X − μX ] = 0

• E[aX + b] = aE[X] + b
 2
• Var [X] = E X − (E[X])2
• If X = a always, then Var [X] = 0

• Var [aX + b] = a2 Var [X]

9
 Uniform Random Variable
• X is a uniform random variable if the PDF of X is
⎧
⎨ 1/(b − a) a ≤ x < b
fX (x) =
⎩ 0 otherwise

where the two parameters are b > a.

• Random Variable Experiment (Virtual Laboratories)

• Quantile Applet (Virtual Laboratories)

10
 Exponential Random Variable
• X is an exponential random variable if the PDF of X is
⎧
⎨ λe−λx x ≥ 0
fX (x) =
⎩ 0 otherwise

where the parameter λ > 0.

• Theorem 3.9: If X is an exponential random variable, then K = X is

a geometric random variable with p = 1 − e−λ .

11
 Erlang Random Variable
• X is an Erlang random variable if the PDF of X is
⎧
⎨ λn xn−1 e−λx x ≥ 0
(n−1)!
fX (x) =
⎩ 0 otherwise

where the parameter λ > 0, and the parameter n ≥ 1 is an integer.

• The n-th order Erlang random variable is the sum of n iid exponential
random variables with parameter λ.

12
 Gaussian RVs
• X is a Gaussian random variable if the PDF of X is
1 −(x−μ)2 /2σ 2
fX (x) = √ e
2πσ 2
where the parameter μ is any real number and the parameter σ > 0.

• Notation: X is N [μ, σ 2 ]
• Properties: E[X] = μ, Var [X] = σ2

13
 Gaussian PDFs
0.8 0.8

0.6 0.6

fX(x)

f (x)
0.4 0.4

X
0.2 0.2

0 0
−2 0 2 4 6 −2 0 2 4 6
x x

(a) μ = 2, σ = 1/2 (b) μ = 2, σ = 2

14
 Linear Transformation
• Theorem 3.13: If X is Gaussian [μ, σ2 ], Y = aX + b is Gaussian
[aμ + b, a2 σ 2 ].

15
 Standard Normal Random Variable
• Z is a standard normal random variable if Z is N [0, 1]
• The CDF of Z is
 z  z
1 −u2 /2
Φ(z) = fZ (u) du = √ e du
−∞ 2π −∞

• Tables (Tables 3.1 and 3.2 on pp. 123 - 124) for Φ(z) and
Q(z) = P [Z > z] = 1 − Φ(z)

• Φ(−z) = 1 − Φ(z)

• Quantile Applet (Virtual Laboratories)

16
 Evaluation of Gaussian CDF
• X is N [μ, σ 2 ], =⇒ Y = aX + b is N [aμ + b, a2 σ 2 ]
X − μX
• Thus Z = is N [0, 1] and
σX
X − μX x − μX
P [X ≤ x] = P ≤
σX σX
or
x − μX x − μX
FX (x) = P Z ≤ =Φ
σX σX

17
 PDFs for Discrete Random Variables
• X discrete: PX (xi )
• In terms of unit step u(x), CDF is

FX (x) = PX (xi ) u(x − xi )
i

• PDF of X is
dFX (x) 
fX (x) = = PX (xi ) δ(x − xi )
dx i

18
 Mixed Random Variables
• Definition 3.14: X is a mixed random variable if and only if fX (x)
contains both impulses and nonzero finite values.

• CDF FX (x) is piecewise cts but has jumps at x1 , x2 , . . .

• Jump at xi is P [X = xi ]
• PDF has impulses at xi weighted by P [X = xi ]

19
 Mixed Random Variables
• Example 3.21
• Problem 3.6.6:
P [line busy] = 0.2
– phone call:
P [no answer] = 0.3
– X = duration of a completed phone call, exponential with E[X] = 3
– W = duration of any call (W = 0 if line busy or no answer)
– Find FW (w) and fW (w)?

20
 Probability Models of Derived Random Variables
• Suppose that Y = g(X). Determine fY (y) from g(X) and fX (x).
• Two step procedure:
1. Find the CDF FY (y) = P [g(X) ≤ y].
2. Compute the PDF by calculating the derivative
fY (y) = dFY (y) /dy.

• Problem 3.7.5: Suppose that U is a uniform over [0, 1] and

X = − ln(1 − U ). Find fX (x).

21
 Probability Models of Derived Random Variables
• Theorem 3.22: Let U be a uniform (0, 1) random variable and let F (x)
denote a cumulative distribution function with an inverse F −1 (u)
defined for 0 < u < 1. The random variable X = F −1 (U ) has CDF
FX (x) = F (x).

• Example 3.27 (MATLAB Demo): U is the uniform (0, 1) random

variable and X = g(U ). Derive g(U ) such that X is the exponential(1)
random variable.

22
 Probability Models of Derived Random Variables
• Theorems 3.19 & 21: If Y = aX + b, a > 0, then Y has CDF and PDF
y−b 1 y−b
FY (y) = FX , fY (y) = fX .
a a a

23
 Conditioning a Continuous Random Variable
• Definition 3.15: For a random variable X with PDF fX (x) and an event
B ⊂ SX with P [B] > 0, the conditional PDF of X given B is
⎧
⎨ fX (x) x ∈ B
⎪
fX|B (x) = P [B]
⎪
⎩ 0 otherwise

• Definition 3.16: Conditional expected value of X given B is

 ∞
E[X|B] = xfX|B (x) dx.
−∞
∞
• E[g(X)|B] = −∞
g(x)fX|B (x) dx

• Conditional expectations: Use fX|B (x).

24
 Virtual Laboratories
• Random Variable Experiment
• Quantile Applet

25
 Chapter Summary
• Continuous random variables, CDF, and PDF fX (x)
• Expected value E[X]

• Mixed random variables

• A function of a random variable Y = g(X)

• Conditional PDF fX|B (x) and conditional expectation E[X|B]

26