MATH 589 Assignment 4: Properties of Symmetric Stable Distributions, Exercises of Mathematics

The problem statement for assignment 4 of a mathematics course (math 589) focusing on the properties of symmetric stable distributions. The assignment includes four problems, covering topics such as showing the necessary and sufficient conditions for a random variable y to have a symmetric stable distribution, proving the absolute continuity of stable distributions, and finding the characteristic function of a random variable yn derived from i.i.d. Uniform random variables.

Typology: Exercises

Pre 2010

Uploaded on 12/23/2021

Ryu4678
Ryu4678 🇺🇸

1 document

1 / 1

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
MATH 589 Assignment 4 Due April 7, 2004
(1) Show that Yhas a symmetric stable distribution iff φY(t)=eb|t|α, where b0 and
02.
(2) Prove that the stable distributions are all absolutely continuous.
(3) For each n1, let Xn
1,Xn
2,... ,Xn
nbe i.i.d. r.v.’s, each having the uniform distribution
on [n, n]. Define
Yn=M
n
m=1
sign Xn
m
|Xn
m|p,
where M>0 and p>1/2. If the Xn
mare positions of masses Mdistributed at random
on [n, n], then Ynis the gravitational force exerted on a unit mass at the origin,
assuming an inverse pth power law.
(a) Show that the c.f. of Ynis
φn(t)=11
n
0
[1 cos(Mt
xp)] dx g(n)n
,
where g(n)0asn→∞.
(b) Show that φn(t)f(t) = exp(
0[1 cos(Mt
xp)] dx)asn→∞.
(c) Make the change of variable y=|Mt|1/px1to show that f(t) is of the form ed|t|α
where d>0 and 0 <α<2.
(d) Show that ed|t|αwhere d>0 and 0 α2 is a c.f.
(4) If λ0 and φ(t) is a c.f. of a random variable, show that eλ(φ(t)1) is an infinitely
divisible c.f.
Hint: for problems 1 and 2, use the formula for the general form of the c.f. of a stable
law.

Partial preview of the text

Download MATH 589 Assignment 4: Properties of Symmetric Stable Distributions and more Exercises Mathematics in PDF only on Docsity!

MATH 589 Assignment 4 Due April 7, 2004

(1) Show that Y has a symmetric stable distribution iff φY (t) = e−b|t|α , where b ≥ 0 and 0 < α ≤ 2.

(2) Prove that the stable distributions are all absolutely continuous.

(3) For each n ≥ 1, let X 1 n , X 2 n ,... , Xnn be i.i.d. r.v.’s, each having the uniform distribution on [−n, n]. Define

Yn = M

∑^ n

m=

sign Xmn |Xmn|p^

where M > 0 and p > 1 /2. If the Xmn are positions of masses M distributed at random on [−n, n], then Yn is the gravitational force exerted on a unit mass at the origin, assuming an inverse pth power law.

(a) Show that the c.f. of Yn is

φn(t) =

n

[∫ ∞

0

[1 − cos(

M t xp^

)] dx − g(n)

])n ,

where g(n) → 0 as n → ∞. (b) Show that φn(t) → f (t) = exp(−

0 [1^ −^ cos(

M t xp^ )]^ dx) as^ n^ → ∞. (c) Make the change of variable y = |M t|^1 /px−^1 to show that f (t) is of the form e−d|t|α where d > 0 and 0 < α < 2. (d) Show that e−d|t|α where d > 0 and 0 ≤ α ≤ 2 is a c.f.

(4) If λ ≥ 0 and φ(t) is a c.f. of a random variable, show that eλ(φ(t)−1)^ is an infinitely divisible c.f.

Hint: for problems 1 and 2, use the formula for the general form of the c.f. of a stable law.