Scribd
Upload
55 views

Video6 - Generating Function

This document provides an overview of generating functions in mathematical methods. It defines the moment generating function (MGF) of a random variable X as the expected value of e raised to the power of t multiplied by X. The MGF can be used to generate non-central moments from the function and its derivatives. The probability generating function (PGF) of an integer-valued random variable Y is defined as the expected value of t raised to the power of Y. PGFs can be used to find probabilities and moments.

Uploaded by

big clap
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
55 views

Video6 - Generating Function

This document provides an overview of generating functions in mathematical methods. It defines the moment generating function (MGF) of a random variable X as the expected value of e raised to the power of t multiplied by X. The MGF can be used to generate non-central moments from the function and its derivatives. The probability generating function (PGF) of an integer-valued random variable Y is defined as the expected value of t raised to the power of Y. PGFs can be used to find probabilities and moments.

Uploaded by

big clap
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Mathematical Methods

Generating Functions

Mathematical methods

School of Risk & Actuarial Studies


UNSW Business School

Video lecture notes

1/5
Mathematical Methods
Generating functions

Moment generating function (mgf) of a r.v.

The moment generating function of a r.v. X is defined as:


h i
MX (t) =E e X ·t
  t2   t3
=1 + E [X ] · t + E X 2 · + E X3 · + ....
2! 3!
Properties of m.g.f.:

Mm·X +b (t) = MX (m · t) · e b·t , for constants m, b;


MX +Y (t) = MX (t) · MY (t), only if X , Y are independent.
x2 x3
*by Taylor series: e x = 1 + x + 2!
+ 3!
+ . . ., with x = X · t.

2/5
Mathematical Methods
Generating functions

Use of moment generating function

Relation m.g.f. and non-central moments: we can write the


m.g.f. as an infinite series of the moments as follows:
∞
h i X tk
MX (t) = E e X ·t = µk · .
k!
k=0

Generating non-central moments using the m.g.f.: we can


generate the moments from the m.g.f. using the relationship:

(r )
µr = E [X r ] = M X (t) .

t=0

3/5
Mathematical Methods
Generating functions

Proof: To prove the above result, consider the continuous


case (similar proof in the discrete case):
Z ∞
(r ) ∂ r h X ·t i ∂r
MX (t) = r E e = r e x·t · fX (x) dx
∂t ∂t −∞
Z ∞ r 
∂ x·t
= r
e · fX (x) dx
−∞ ∂t
Z ∞
x r · e x·t · fX (x) dx

=
−∞
h i
=E X r · e X ·t .

Set t = 0 and you get the desired result.


Remark: If the m.g.f. exists for t in an open interval containing zero,
then it uniquely determines the probability distribution.

4/5
Mathematical Methods
Generating functions

Probability generating function (p.g.f.) of a r.v.

Let Y be an integer-valued random variable with


Pr(Y = i) = pi for i = 0, 1, 2, . . ., the p.g.f. is defined as:
h i X ∞
PY (t) = E t Y = pY (i) · t i .
i=1
Properties of p.g.f.:
- The relationship between p.g.f. and m.g.f. is as follows:

PY (t) = MY (log(t)) .

(r )
- Probabilities: Pr(Y = r ) = P Y (t) /r !

t=0
- Take the k thderivative and set t = 1:
(k)
PY (1) = E Y · (Y − 1) · (Y − 2) · . . . · (Y − k + 1) · 1Y −k .


5/5

You might also like