Chapter 11

Characteristic Functions

[33/100%]
Show
LaTeX
Problem 11.1

Find the characteristic function of a random variable XX such that P({X=1})=P({X=−1})=12\mathbb {P}\left(\{ X=1\} \right)=\mathbb {P}\left(\{ X=-1\} \right)=\frac{1}{2}.

?
Problem 11.2

Find the characteristic function of a random variable XX such that P({X=1})=p\mathbb {P}\left(\{ X=1\} \right)=p and P({X=0})=q\mathbb {P}\left(\{ X=0\} \right)=q, where p+q=1p+q=1 (Bernoulli distribution).

?
Problem 11.3

Find the characteristic function of a random variable XX such that P({X=k})=2−k,k=1,2,…\mathbb {P}\left(\{ X=k\} \right)=2^{-k}, k=1,2, \ldots.

?
Problem 11.4

Find the characteristic function of a random variable XX with Poisson distribution. Can you sketch the graph of ϕX(t)\phi_{X}(t) ?

?
Problem 11.5

Find the characteristic function of a random variable XX with uniform distribution over the interval [−1,1][-1,1]. Sketch the graph of this characteristic function.

?
Problem 11.6

Find the characteristic function ϕX(t)\phi_{X}(t) of a random variable XX having distribution with density

fX(x)={1−∣x∣ if ∣x∣⩽10 otherwise  f_{X}(x)= \begin{cases} 1-\left|x\right| & \text{ if }\left|x\right| \leqslant 1 \\ 0 & \text{ otherwise } \end{cases}
?
Problem 11.7

Find the characteristic function ϕX(t)\phi_{X}(t) of a random variable XX with distribution

PX(A)=12  1A(0)+14∫−11  1A(x)dx P_{X}(A)=\frac{1}{2} \; \mathbb {1}_{A}(0)+\frac{1}{4} \int _{-1}^{1} \; \mathbb {1}_{A}(x) \mathrm{d} x
?
Problem 11.8

What is the characteristic function ϕX(t)\phi_{X}(t) of a random variable XX with exponential distribution? Sketch the graph of ϕX(t)\phi_{X}(t).

?
Problem 11.9

What is the characteristic function ϕX(t)\phi_{X}(t) of a normally distributed random variable XX with mean 0 and variance 1 ?

?
Problem 11.10

Show that for any random variable XX

?
(a)

ϕX(0)=1\phi_{X}(0)=1,

(b)

∣ϕX(t)∣⩽1\left|\phi_{X}(t)\right| \leqslant 1 for all t∈Rt \in \mathbb {R}.

(c)

ϕX(−t)=ϕX(t)‾\phi_{X}(-t)=\overline{\phi_{X}(t)} for all t∈Rt \in \mathbb {R}. (By zˉ\bar{z} we denote the complex conjugate of z∈Cz \in \mathbb {C}; that is, if z=a+ibz=a+i b, where a,b∈Ra, b \in \mathbb {R}, then zˉ=a−ib\bar{z}=a-i b.)

Problem 11.11

Let XX and YY be random variables such that Y=aX+bY=a X+b, where a,b∈Ra, b \in \mathbb {R}. Show that

ϕY(t)=eitbϕX(at) \phi _{Y}(t)=e^{i t b} \phi _{X}(a t)
?
Problem 11.12

Use the formula in Problem 11.11 and the characteristic function of the normal distribution with mean 0 and variance 1 obtained in Problem 11.9 to find the characteristic function of the normal distribution with arbitrary mean mm and variance σ2>0\sigma^{2}>0.

?
Problem 11.13

Let XX be a random variable such that P({X∈Z})=1\mathbb {P}\left(\{ X \in \mathbb {Z}\} \right)=1. Demonstrate that ϕX(t)\phi_{X}(t) is a periodic function with period 2π2 \pi; that is, ϕX(t+2π)=ϕX(t)\phi_{X}(t+2 \pi )=\phi_{X}(t) for all tt.

?
Problem 11.14

Is it true that if ϕX(2π)=1\phi_{X}(2 \pi )=1, then P({X∈Z})=1\mathbb {P}\left(\{ X \in \mathbb {Z}\} \right)=1 ?

?
Problem 11.15

Let XX be a random variable and let a≠0a \neq 0. Show that the following conditions are equivalent:

?
(a)

ϕX(a)=1\phi_{X}(a)=1;

(b)

ϕX\phi_{X} is periodic with period ∣a∣\left|a\right|;

(c)

P({a2πX∈Z})=1P\left(\left\{ \frac{a}{2 \pi } X \in \mathbb {Z}\right\} \right)=1.

Problem 11.16

If XX is a random variable with density, then ϕX(t)<1\phi_{X}(t)<1 for all t≠0t \neq 0.

?
Problem 11.17

Let XX be a random variable taking values in a finite set. Verify that the moments E[Xn]\mathbb {E}\left[X^{n}\right] of XX are given by

E[Xn]=1indndtnϕX(0) \mathbb {E}\left[X^{n}\right]=\frac{1}{i^{n}} \frac{d^{n}}{d t^{n}} \phi _{X}(0)
?
Problem 11.18

Use the formula in Theorem 11.1 to obtain an expression for the variance of XX in terms of the characteristic function ϕX\phi_{X} and its derivatives at zero.

?
Problem 11.19

Use Theorem 11.1 to compute the expectation E[X]\mathbb {E}\left[X\right] and the second moment E[X2]\mathbb {E}\left[X^{2}\right] of a random variable XX with Poisson distribution.

?
Problem 11.20

Use Theorem 11.1 to compute the expectation E[X]\mathbb {E}\left[X\right] and the second moment E[X2]\mathbb {E}\left[X^{2}\right] of a random variable XX with normal distribution.

?
Problem 11.21

Let XX be a random variable such that P({X∈Z})=1\mathbb {P}\left(\{ X \in \mathbb {Z}\} \right)=1. Show that for all n∈Zn \in \mathbb {Z},

P({X=n})=12π∫02πe−itnϕX(t)dt \mathbb {P}\left(\{ X=n\} \right)=\frac{1}{2 \pi } \int _{0}^{2 \pi } e^{-i t n} \phi _{X}(t) \mathrm{d} t
?
Problem 11.22

What is the distribution of a random variable XX with characteristic function

ϕX(t)=12e−it+13+16e2it? \phi _{X}(t)=\frac{1}{2} e^{-i t}+\frac{1}{3}+\frac{1}{6} e^{2 i t} ?
?
Problem 11.23

Find the distribution of a random variable XX with characteristic function

ϕX(t)=cos⁡(t2) \phi _{X}(t)=\cos \left(\frac{t}{2}\right)
?
Problem 11.24

Find the distribution of a random variable XX with characteristic function

ϕX(t)=23eit−1 \phi _{X}(t)=\frac{2}{3 e^{i t}-1}
?
Problem 11.25

Find the distribution of a random variable with characteristic function e−∣t∣e^{-\left|t\right|}.

?
Problem 11.26

What is the distribution of a random variable with characteristic function e−t2/2e^{-t^{2} / 2} ?

?
Problem 11.27

Find the distribution of a random variable with characteristic function 11+t2\frac{1}{1+t^{2}}.

?
Problem 11.28

Find the distribution of a random variable with characteristic function 13eit2+23eit3\frac{1}{3} e^{i t \sqrt{2}}+\frac{2}{3} e^{i t \sqrt{3}}.

?
Problem 11.29

What is the distribution of a random variable with characteristic function 12e−t2/2+12\frac{1}{2} e^{-t^{2} / 2}+\frac{1}{2} ?

?
Problem 11.30

Prove Theorem 11.4.

?
Problem 11.31

Use Theorem 11.4 to find the characteristic function of a random variable with the binomial distribution.

?
Problem 11.32

Let XX and YY be independent random variables with Poisson distribution. Verify that X+YX+Y is also a random variable with Poisson distribution. If λ\lambda and μ\mu are the parameters of XX and YY, respectively, what is the parameter of X+YX+Y ?

?
Problem 11.33

Let XX and YY be normally distributed independent random variables. Show that X+YX+Y is also normally distributed. If XX has mean mm and variance σ2\sigma^{2} and YY has mean m~\tilde{m} and variance σ~2\tilde{\sigma }^{2}, what is the mean and variance of X+YX+Y ?

?