Scribd
Upload
3 views1 page

Sec 01

Uploaded by

Sambsamb Sambiani
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views1 page

Sec 01

Uploaded by

Sambsamb Sambiani
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Large Sample Theory

Ferguson

Exercises, Section 1, Modes of Convergence.


1. (a) The case r = 1 of Chebyshev’s Inequality is known as Markov’s Inequality and
is usually written P(|X| ≥ ) ≤ E(|X|)/ for an arbitrary random variable X and  > 0.
For each fixed  > 1, find a distribution for X with EX = 0 and E|X| = 1 that gives
equality in Markov’s inequality.
(b) Prove for an arbitrary random variable X,

E cosh(X) − 1
P(|X| ≥ ) ≤ .
cosh() − 1

2. Suppose that fn (x) and g(x) are densities, and that for all x, fn (x) converges to
some constant (independent of x) times g(x). Does it follow that the random variable with
density fn (x) converges in law to the random variable with density g(x)? If so, show it. If
not, give a counterexample.
3. Suppose that Xn has the binomial distribution, B(n, p) for some 0 < p < 1.
(a) For fixed k, find limn→∞ P(Xn ≤ k − 1|Xn ≤ k).
(b) Let the distribution of Yn be the conditional distribution of Xn given Xn ≤ k.
L
Express the result in (a) in the form Yn −→ Y .
4. Let X have an inverse power distribution with distribution function, FX (x) =
1 − x−α for x ∈ [1, ∞), where α > 0.
(a) Show EX = α/(α − 1) for α > 1, and Var(X) = α/((α − 2)(α − 1)2 ) for α > 2.
(b) Let Y = (α − 1)X − α (so that Y has mean 0 and variance tending to 1 as α → ∞.
L
Show that Y −→ Z as α → ∞ for some random variable Z and find the distribution
function of Z.
P
5. We say Xn −→ ∞ if for every number B (no matter how big), P(Xn > B) → 1
a.s.
as n → ∞. We say Xn −→ ∞ if for every number B, P(Xk > B for every k ≥ n) → 1 as
n → ∞.
Suppose X1 , X2 , . . . are independent with P(0 ≤ Xj ≤ 1) = 1 for all j. Let Sn =
n P
1 Xj and µn = ESn . Show that if µn → ∞ as n → ∞, then Sn −→ ∞. (Hint: Show
a.s.
Var(Xi ) ≤ EXi and use Chebyshev’s inequality.) Show that this implies Sn −→ ∞.

You might also like