Solutions of Selected Problems from Probability

Essentials, Second Edition

Solutions to selected problems of Chapter 2
2.1 Lets first prove by induction that #(2n ) = 2n if = {x1 , . . . , xn }. For n = 1
it is clear that #(21 ) = #({, {x1 }}) = 2. Suppose #(2n1 ) = 2n1 . Observe that
2n = {{xn } A, A 2n1 } 2n1 } hence #(2n ) = 2#(2n1 ) = 2n . This proves
finiteness. To show that 2 is a -algebra we check:
1. hence 2 .
2. If A 2 then A and Ac hence Ac 2S
.
3. Let (An )n1 be a sequence of subsets of . Then
n=1 An is also a subset of hence

in 2 .
Therefore 2 is a -algebra.
2.2 We check if H = A G has the three properties of a -algebra:
1. G A hence A G .
2. If B A G then B G A. This implies that B c G A since each
G is a -algebra. So B c A G .
S
3. Let (An )n1 be a sequence S
in H. Since each An G ,
n=1 An is in G since G is a

-algebra for each A. Hence n=1 An A G .

Therefore H = A G is a -algebra.
c
c

c
2.3 a. Let x (
n=1 An ) . Then x An for all n, hence x n=1 An . So (n=1 An )
c

c
c

n=1 An . Similarly if x n=1 An then x An for any n hence x (n=1 An ) . So

c
c

(n=1 An ) = n=1 An .

c c

c
b. By part-a
n=1 An = (n=1 An ) , hence (n=1 An ) = n=1 An .

2.4 lim inf n An =

n=1 Bn where Bn = mn Am A n since A is closed under taking
countable intersections. Therefore lim inf n An A since A is closed under taking
countable unions.
By De Morgans Law it is easy to see that lim sup An = (lim inf n Acn )c , hence lim supn An
A since lim inf n Acn A and A is closed under taking complements.
Note that x lim inf n An n s.t x mn Am x mn Am n x
lim supn An . Therefore lim inf n An lim supn An .
2.8 Let L = {B R : f 1 (B) B}. It is easy to check that L is a -algebra. Since f
is continuous f 1 (B) is open (hence Borel) if B is open. Therefore L contains the open
sets which implies L B since B is generated by the open sets of R. This proves that
f 1 (B) B if B B and that A = {A R : B B with A = f 1 (B) B} B.

Solutions to selected problems of Chapter 3

3.7 a. Since P (B) > 0 P (.|B) defines a probability measure on A, therefore by Theorem
2.4 limn P (An |B) = P (A|B).
b. We have that A Bn A B since 1ABn (w) = 1A (w)1Bn (w) 1A (w)1B (w).
Hence P (A Bn ) P (A B). Also P (Bn ) P (B). Hence
P (A|Bn ) =

P (A Bn )
P (A B)

= P (A|B).
P (Bn )
P (B)

c.
P (An Bn )
P (A B)

= P (A|B)
P (Bn )
P (B)
since An Bn A B and Bn B.
P (An |Bn ) =

3.11 Let B = {x1 , x2 , . . . , xb } and R = {y1 , y2 , . . . , yr } be the sets of b blue balls and r
red balls respectively. Let B 0 = {xb+1 , xb+2 , . . . , xb+d } and R0 = {yr+1 , yr+2 , . . . , yr+d } be
the sets of d-new blue balls and d-new red balls respectively. Then we can write down the
sample space as
= {(a, b) : (a B and b B B 0 R) or (a R and b R R0 B)}.
Clearly card() = b(b + d + r) + r(b + d + r) = (b + r)(b + d + r). Now we can define a
probability measure P on 2 by
P (A) =

card(A)
.
card()

a. Let
A = { second ball drawn is blue}
= {(a, b) : a B, b B B 0 } {(a, b) : a R, b B}
card(A) = b(b + d) + rb = b(b + d + r), hence P (A) =
b. Let

b
.
b+r

B = { first ball drawn is blue}

= {(a, b) : a B}
Observe A B = {(a, b) : a B, b B B 0 } and card(A B) = b(b + d). Hence
P (B|A) =

card(A B)
b+d
P (A B)
=
=
.
P (A)
card(A)
b+d+r
2

3.17 We will use the inequality 1 x > ex for x > 0, which is obtained by taking Taylors
expansion of ex around 0.
P ((A1 . . . An )c ) = P (Ac1 . . . Acn )
= (1 P (A1 )) . . . (1 P (An ))
exp(P (A1 )) . . . exp(P (An )) = exp(

n
X
i=1

P (Ai ))

Solutions to selected problems of Chapter 4

4.1 Observe that
 k
nk

n
P (k successes) =
1
2
n
n
= Can b1,n . . . bk,n dn


where

nj+1

k
an = (1 )n bj,n =
dn = (1 )k
k!
n
n
n
1 j and dn 1 as n . Observe that

C=
It is clear that bj,n

log((1

n
2 1
) ) = n( 2 2 ) for some (1 n , 1)
n
n n

by Taylor series expansion of log(x) around 1. It follows that an e as n and

that

2 1
|Error| = |en log(1 n ) e | |n log(1 ) | = n 2 2 p
n
n
Hence in order to have a good approximation we need n large and p small as well as to
be of moderate size.

Solutions to selected problems of Chapter 5

5.7 We put xn = P (X is even) for X B(p, n). Let us prove by induction that xn =
1
(1 + (1 2p)n ). For n = 1, x1 = 1 p = 21 (1 + (1 2p)1 ). Assume the formula is true for
2
n 1. If we condition on the outcome of the first trial we can write
xn = p(1 xn1 ) + (1 p)xn
1
1
= p(1 (1 + (1 2p)n1 )) + (1 p)( (1 + (1 2p)n1 ))
2
2
1
=
(1 + (1 2p)n )
2
hence we have the result.
P
P
P
5.11 Observe that E(|X |) = i< ( i)pi + i (i )pi . Since i (i )pi =
P
P
P
i=0 (i )pi
i< (i )pi we have that E(|X |) = 2
i< ( i)pi . So
X
E(|X |) = 2
( i)pi
i<

= 2

1
X

( i)

i=1

e k
k!

= 2e

1
X
k
k+1

)
(
k!
(k 1)!
i=0

= 2e

.
(k 1)!

Solutions to selected problems of Chapter 7

7.1 Suppose limn P (An ) 6= 0. Then there exists  > 0 such P
that there are dis
tinct An1 , An2 , . . . with P (Ank ) > 0 for every k 1. This gives
k=1 P (Ank ) =
which
since by the hypothesis that the An are disjoint we have that
P is a contradiction

P
(A
)
=
P
(
A
nk
n=1 nk ) 1 .
k=1
7.2 Let An = {A : P (A ) > 1/n}. An is a finite set P
otherwise we can pick disjoint

A1 , A2 , . . . in An . This would give us P m=1 Am =

m=1 P (Am ) = which is a
contradiction. Now {A : B} =
A
hence
(A
)
B is countable since it is a
n=1 n
countable union of finite sets.
7.11 Note that {x0 } =
n=1 [x0 1/n, x0 ] therefore {x0 } is a Borel set. P ({x0 }) =
limn P ([x0 1/n, x0 ]). Assuming that f is continuous we have that f is bounded
by some M on the interval [x0 1/n, x0 ] hence P ({x0 }) = limn M (1/n) = 0.
Remark: In order this result to be true we dont need f to be continuous. When we define
the Lebesgue integral (or more generally integral with respect to a measure) and study its
properties we will see that this result is true for all Borel measurable non-negative f .
7.16 First observe that F (x) F (x) > 0 iff P ({x}) > 0. The family of events {{x} :
P ({x}) > 0} can be at most countable as we have proven in problem 7.2 since these events
are disjoint and have positive probability. Hence F can have at most countable discontinuities. For an example with infinitely many jump discontinuities consider the Poisson
distribution.

7.18 Let F be as given. It is clear that F is a nondecreasing function. For x < 0 and x 1
right continuity of F is clear. For any 0 < x < 1 let i be such that i1+1 x < i1 . If
xn x then there exists N such that i1+1 xn < i1 for every n N . Hence F (xn ) = F (x)
for every n N which implies that F is right continuous P
at x. For x = 0 we have that
1
F (0) = 0. Note that for any  there exists N such that
i=N 2i < . So for all x s.t.
|x| N1 we have that F (x) . Hence F (0+) = 0. This proves the right continuity of F
P
1
for all x. We also have that F () =
i=1 2i = 1 and F () = 0 so F is a distribution
function of a probability on R.
P
1
1
a. P ([1, )) = F () F (1) = 1
n=2 = 1 2 = 2 .
P

1
1
1
b. P ([ 10 , )) = F () F ( 10 ) = 1 n=11 2i = 1 210 .
c P ({0}) = F (0) F (0) = 0. P
1
1
d. P ([0, 21 )) = F ( 21 ) F (0) =
n=3 2i 0 = 4 .
e. P ((, 0)) = F (0) = 0.
f. P ((0, )) = 1 F (0) = 1.
6

Solutions to selected problems of Chapter 9

9.1 It is clear by the definition of F that X 1 (B) F for every B B. So X is measurable
from (, F) to (R, B).
9.2 Since X is both F and G measurable for any B B, P (X B) = P (X B)P (X
B) = 0 or 1. Without loss of generality we can assume that there exists a closed interval
I such that P (I) = 1. Let n = {tn0 , . . . tnln } be a partition of I such that n n+1 and
supk tnk tnk1 0. For each n there exists k (n) such that P (X [tnk , tnk +1 ]) = 1 and
[tnk (n+1 , tnk (n+1)+1 ] [tnk (n) , tnk (n)+1 ]. Now an = tnk (n) and bn = tnk (n) + 1 are both Cauchy
sequences with a common limit c. So 1 = limn P (X (tnk , tnk +1 ]) = P (X = c).
9.3 X 1 (A) = (Y 1 (A) (Y 1 (A) X 1 (A)c )c )(X 1 (A) Y 1 (A)c ). Observe that both
Y 1 (A) (X 1 (A))c and X 1 (A) Y 1 (A)c are null sets and therefore measurable. Hence
if Y 1 (A) A0 then X 1 (A) A0 . In other words if Y is A0 measurable so is X.
9.4 Since X is integrable, for any  > 0 there exists M such that
the dominated convergence theorem. Note that

|X|1{X>M } dP <  by

E[X1An ] = E[X1An 1{X>M } ] + E[X1An 1{XM } ]

E[|X|1{XM } ] + M P (An )
Since P (An ) 0, there exists N such that P (An )
E[X1An ]  +  n N , i.e. limn E[X1An ] = 0.


M

for every n N . Therefore

9.5 It is clear that 0 Q(A) 1 and Q() = 1 since X is nonnegative and E[X] = 1. Let
A1 , A2 , . . . be disjoint. Then
Q(
n=1 An )

X
X
= E[X1
] = E[
X1An ] =
E[X1An ]
n=1 An
n=1

n=1

where the last equality follows from the monotone convergence theorem. Hence Q(
n=1 An ) =
P

Q(A
).
Therefore
Q
is
a
probability
measure.
n
n=1
9.6 If P (A) = 0 then X1A = 0 a.s. Hence Q(A) = E[X1A ] = 0. Now assume P is the
uniform distribution on [0, 1]. Let X(x) = 21[0,1/2] (x). Corresponding measure Q assigns
zero measure to (1/2, 1], however P ((1/2, 1]) = 1/2 6= 0.
9.7 Lets prove this first for simple functions, i.e. let Y be of the form
Y =

n
X
i=1

ci 1Ai

for disjoint A1 , . . . , An . Then

EQ [Y ] =

n
X

ci Q(Ai ) =

i=1

n
X

ci E[X1Ai ] = EP [XY ]

i=1

For non-negative Y we take a sequence of simple functions Yn Y . Then

EQ [Y ] = lim EQ [Yn ] = lim EP [XYn ] = EP [XY ]
n

where the last equality follows from the monotone convergence theorem. For general Y
L1 (Q) we have that EQ [Y ] = EQ [Y + ] EQ [Y ] = EP [(XY )+ ] EQ [(XY ) ] = EP [XY ].
9.8 a. Note that X1 X = 1 a.s. since P (X > 0) = 1. By problem 9.7 EQ [ X1 ] = EP [ X1 X] = 1.
So X1 is Q-integrable.
b. R : A R, R(A) = EQ [ X1 1A ] is a probability measure since X1 is non-negative and
EQ [ X1 ] = 1. Also R(A) = EQ [ X1 1A ] = EP [ X1 X1A ] = P (A). So R = P .
9.9 Since P (A) = EQ [ X1 1A ] we have that Q(A) = 0 P (A) = 0. Now combining the
results of the previous problems we can easily observe that Q(A) = 0 P (A) = 0 iff
P (X > 0) = 1.
9.17. Let

((x )b + )2
.
2 (1 + b2 )2
Observe that {X + b} {g(X) 1}. So
g(x) =

P ({X + b}) P ({g(X) 1})

E[g(X)]
1

where the last inequality follows from Markovs inequality. Since E[g(X)] =
get that
1
.
P ({X + b})
1 + b2
9.19
xP ({X > x}) E[X1{ X > x}]
Z
z2
z
e 2 dz
=
2
x
x2

e 2
=
2
Hence

x2

e 2
P ({X > x})
x 2
9

2 (1+b2 )
2 (1+b2 )2

we

.
9.21 h(t + s) = P ({X > t + s}) = P ({X > t + s, X > s}) = P ({X > t + s|X > s})P ({X >
1
m
s}) = h(t)h(s) for all t, s > 0. Note that this gives h( n1 ) = h(1) n and h( m
) = h(1) n . So
n
for all rational r we have that h(r) = exp (log(h(1))r). Since h is right continuous this
gives h(x) = exp(log(h(1))x) for all x > 0. Hence X has exponential distribution with
parameter log h(1).

10

Solutions to selected problems of Chapter 10

10.5 Let P be the uniform distribution on [1/2, 1/2]. Let X(x) = 1[1/4,1/4] and Y (x) =
1[1/4,1/4]c . It is clear that XY = 0 hence E[XY ] = 0. It is also true that E[X] = 0. So
E[XY ] = E[X]E[Y ] however it is clear that X and Y are not independent.
10.6 a. P (min(X, Y ) > i) = P (X > i)P (Y > i) = 21i 21i = 41i . So P (min(X, Y ) i) =
1 P (min(X, Y ) > i) = 1 41i .
P
P 1 1
1
1
b. P (X = Y ) =
i=1 P (X = i)P (Y = i) =
i=1 2i 2i = 1 1i 1 = 3 .
4
P
P
1 1
c. P (Y > X) =
(Y > i)P (X = i) =
= 13 .
i=1 P
2i 2i
P P 1 1
P i=1
d. P (X divides Y ) = i=1 k=1 2i 2ki = i=1 21i 2i11 .
P
P 1 1
2
e. P (X kY ) =
P
(X

ki)P
(Y
=
i)
=
i=1
i=1 2i 2ki 1 = 2k+1 1 .

11

Solutions to selected problems of Chapter 11

11.11. Since P {X > 0} = 1 we have that P {Y < 1} = 1. So FY (y) = 1 for y 1. Also
P {Y 0} = 0 hence FY (y) = 0 for y 0. For 0 < y < 1 P {Y > y} = P {X < 1y
}=
y
1y
FX ( y ). So
Z 1y
Z y
y
1
1z
FY (y) = 1
fX (x)dx = 1
)dz
f
(
X
2
z
0 z
0
by change of variables. Hence

< y 0
0
1y
1
0<y1
fY (y) =
2 fX ( y )
y
0
1y<
11.15 Let G(u) = inf{x : F (x) u}. We would like to show {u : G(u) > y} = {u :
F (Y ) < u}. Let u be such that G(u) > y. Then F (y) < u by definition of G. Hence
{u : G(u) > y} {u : F (Y ) < u}. Now let u be such that F (y) < u. Then y < x for any x
such that F (x) u by monotonicity of F . Now by right continuity and the monotonicity of
F we have that F (G(u)) = inf F (x)u F (x) u. Then by the previous statement y < G(u).
So {u : G(u) > y} = {u : F (Y ) < u}. Now P {G(U ) > y} = P {U > F (y)} = 1 F (y) so
G(U ) has the desired distribution. Remark:We only assumed the right continuity
of F .

12

Solutions to selected problems of Chapter 12

2

XY
2
12.6 Let Z = ( 1Y )Y ( XY
)X. Then Z2 = ( 12 )Y2 ( XY
2 )X 2( )Cov(X, Y ) =
X
X Y
Y
X
1 2XY . Note that XY = 1 implies Z2 = 0 which implies Z = c a.s. for some constant
c. In this case X = YXXY (Y c) hence X is an affine function of Y .

p
12.11 Consider the mapping g(x, y) = ( x2 + y 2 , arctan( xy )). Let S0 = {(x, y) : y = 0},
S1 = {(x, y) : y > 0}, S2 = {(x, y) : y < 0}. Note that 2i=0 Si = R2 and m2 (S0 ) = 0.
Also for i = 1, 2 g : Si R2 is injective and continuously differentiable. Corresponding
inverses are given by g11 (z, w) = (z sin w, z cos w) and g21 (z, w) = (z sin w, z cos w). In
both cases we have that |Jgi1 (z, w)| = z hence by Corollary 12.1 the density of (Z, W ) is
given by
1 z2
1 z2
2 z +
fZ,W (z, w) = (
e
e 2 z)1( 2 , 2 ) (w)1(0,) (z)
2 2
2 2
1
z z2
=
1( 2 , 2 ) (w) 2 e 2 1(0,) (z)

as desired.
12.12 Let P be the set of all permutations of {1, . . . , n}. For any P let X be the
corresponding permutation of X, i.e. Xk = Xk . Observe that
P (X1 x1 , . . . , Xn xn ) = F (x1 ) . . . F (Xn )
hence the law of X and X coincide on a system generating B n therefore they are equal.
Now let 0 = {(x1 , . . . , xn ) Rn : x1 < x2 < . . . < xn }. Since Xi are i.i.d and have
continuous distribution PX (0 ) = 1. Observe that
P {Y1 y1 , . . . , Yn yn } = P (P {X1 y1 , . . . , Xn yn } 0 )
Note that {X1 y1 , . . . , Xn yn } 0 , P are disjoint and P (0 = 1) hence
X
P {X1 y1 , . . . , Xn yn }
P {Y1 y1 , . . . , Yn yn } =
P

= n!F (y1 ) . . . F (yn )

for y1 . . . yn . Hence

fY (y1 , . . . , yn ) =

n!f (y1 ) . . . f (yn ) y1 . . . yn

0
otherwise

13

Solutions to selected problems of Chapter 14

14.7 X (u) is real valued iff X (u) = X (u) = X (u). By uniqueness theorem X (u) =
X (u) iff FX = FX . Hence X (u) is real valued iff FX = FX .
14.9
that
is true for n = 1. Put Yn =
Pn We use induction. It is clear
Pn the statement
d3
3
3
i=1 Xi and assume that E[(Yn ) ] =
i=1 E[(Xi ) ]. Note that this implies dx3 Yn (0) =
P
d3
i ni=1 E[(Xi )3 ]. Now E[(Yn+1 )3 ] = E[(Xn+1 + Yn )3 ] = i dx
3 (Xn+1 Yn )(0) by independence of Xn+1 and Yn . Note that
d3
d3

(0)
=
X (0)Yn (0)
X
Y
n
n+1
dx3
dx3 n+1
d
d2
d
d2
+ 3 2 Xn+1 (0) Yn (0) + 3 Xn+1 (0) 2 Yn (0)
dx
dx
dx
dx
3
d
+ Xn+1 (0) 3 Yn (0)
dx
d3
d3
=

Y (0)
(0)
+
X
dx3 n+1
dx3 n
!
n
X
= i E[(Xn+1 )3 ] +
E[(Xi )3 ]
i=1
d
where we used the fact that dx
Xn+1 (0) = iE(Xn+1 ) = 0 and
P
n+1
3
3
E[(Yn+1 ) ] = i=1 E[(Xi ) ] hence the induction is complete.

14.10 It is clear that 0 (A) 1 since

0

n
X

j j (A)

j=1

n
X

j = 1.

j=1

Also for Ai disjoint

(
i=1 Ai )

n
X
j=1
n
X

j j (
i=1 Ai )
j

j=1
i=1

n
XX

j j (Ai )

i=1 j=1

j (Ai )

X
i=1

14

(Ai )

d
(0)
dx Yn

= iE(Yn ) = 0. So

R
Hence
Pn Ris countably additive therefore it is a probability mesure. Note that 1A d(dx) =
1A (x)dj (dx) by definition of . Now by linearity and monotone convergence
j=1 j
R
R
P
theorem for a non-negative Borel function f we have that f (x)(dx) = nj=1 j f (x)dj (dx).
R
R
P
Extending this to integrable f we have that (u) = eiux (dx) = nj=1 j eiux dj (dx) =
Pn
j (u).
j=1 j
14.11 Let be the double exponential distribution, 1 be the distribution of Y and 2 be
the distribution of Y
R where Y is anRexponential r.v. with parameter = 1. Then we
have that (A) = 21 A(0,) ex dx + 21 A(,0) ex dx = 12 1 (A) + 21 2 (A). By the previous
1
1
1
exercise we have that (u) = 21
1 (u) + 12
2 (u) = 12 ( 1iu
+ 1+iu
) = 1+u
2.
n

d
s /2
14.15. Note that E{X n } = (i)n dx
. Note that
n X (0). Since X N (0, 1) X (s) = e
s2 /2
we can get the derivatives of any order of e
at 0 simply by taking Taylors expansion
x
of e :

X
(s2 /2)n
2
es /2 =
n!
i=0

X
1 (i)2n (2n)! 2n
s
=
n n!
2n!
2
i=0
n

2k

d
d
2k
hence E{X n } = (i)n dx
} = (i)2k dx
n X (0) = 0 for n odd. For n = 2k E{X
2k X (0) =
2k (2k)!
(i)
(2k)!
(i)2k 2k k! = 2k k! as desired.

15

Solutions to selected problems of Chapter 15

P
15.1 a. E{x} = n1 ni=1 E{Xi } = .
P
2
b. Since X1 , . . . , Xn are independent Var(x) = n12 ni=1 Var{Xi } = n .
P
P
2
c. Note that S 2 = n1 ni=1 (Xi )2 x2 . Hence E(S 2 ) = n1 ni=1 ( 2 + 2 ) ( n + 2 ) =
n1 2
.
n
Q

15.17 Note that Y (u) =

i=1 Xi (u) = ( iu ) which is the characteristic function
of Gamma(,) random variable. Hence by uniqueness of characteristic function Y is
Gamma(,).

16

Solutions to selected problems of Chapter 16

16.3 P ({Y y}) = P ({X y} {Z = 1}) + P ({X y} {Z = 1}) = 12 (y) +
1
(y) = (y) since Z and X are independent and (y) is symmetric. So Y is normal.
2
Note that P (X + Y = 0) = 21 hence X + Y can not be normal. So (X, Y ) is not Gaussian
even though both X and Y are normal.
16.4 Observe that

Q = X Y

X
Y

Y
X

So det(Q) = X Y (1 2 ). So det(Q) = 0 iff = 1. By Corollary 16.2 the joint density

of (X, Y ) exists iff 1 < < 1. (By Cauchy-Schwartz we know that 1 1). Note
that
Y

1
X
Q1 =

X Y (1 2 ) XY
Substituting this in formula 16.5 we get that
(


2
1
x X
2(1 2 )
X
!)

2
2(x X )(y Y )
y Y

+
.
X Y
Y

1
exp
f(X,Y ) (x, y) =
2X Y (1 2 )

16.6 By Theorem 16.2 there exists a multivariate normal r.v. Y with E(Y ) = 0 and a
diagonal covariance matrix s.t. X = AY where A is an orthogonal matrix. Since
Q = AA and det(Q) > 0 the diagonal entries of are strictly positive hence we can
of B(X ) is given by
define B = 1/2 A . Now the covariance matrix Q
= 1/2 A AA A1/2
Q
= I
So B(X ) is standard normal.
16.17 We know that as in Exercise 16.6 if B = 1/2 A where A is the orthogonal matrix s.t.
Q = AA then B(X ) is standard normal. Note that this gives (X ) Q1 (X ) =
(X ) B B(X ) which has chi-square distribution with n degrees of freedom.

17

Solutions to selected problems of Chapter 17

17.1 Let n(m) and j(m) be such that Ym = n(m)1/p Zn(m),j(m) . This gives that P (|Ym | >
1
0 as m . So Ym converges to 0 in probability. However E[|Ym |p ] =
0) = n(m)
E[n(m)Zn(m),j(m) ] = 1 for all m. So Ym does not converge to 0 in Lp .
17.2 Let Xn = 1/n. It is clear that Xn converge to 0 in probability. If f (x) = 1{0} (x) then
we have that P (|f (Xn ) f (0)| > ) = 1 for every  1, so f (Xn ) does not converge to
f (0) in probability.
P
P
17.3 First observe that E(Sn ) = ni=1 E(Xn ) = 0 and that Var(Sn ) = ni=1 Var(Xn ) = n
since E(Xn ) = 0 and Var(Xn ) = E(Xn2 ) = 1. By Chebyshevs inequality P (| Snn | ) =
P (|S | n) Var(Sn ) = n 0 as n . Hence Sn converges to 0 in probability.
n

n2 2

n2 2

Sn2
|
n2

P 1
1
) n2
.
Since
2
i=1 n2 2 < by

c
Sn2
S

Borel Cantelli Theorem P (lim supn {| n2 | }) = 0. Let 0 = m=1 lim supn {| nn22 | m1 } .
17.4 Note that Chebyshevs inequality gives P (|

Then P (0 ) = 1. Now lets pick w 0 . For any  there exists m s.t. m1  and
S
S
w (lim supn {| nn22 | m1 })c . Hence there are finitely many n s.t. | nn22 | m1 which implies
S
S (w)
that there exists N (w) s.t. | nn22 | m1 for every n N (w). Hence nn2 2 0. Since
P (0 ) = 1 we have almost sure convergence.
17.12 Y < a.s. which follows by Exercise 17.11 since Xn < and X < a.s. Let
1
Z = 1c 1+Y
. Observe that Z > 0 a.s. and EP (Z) = 1. Therefore as in Exercise 9.8
Q(A) = EP (Z1A ) defines a probability measure and EQ (|Xn X|) = EP (Z|Xn X|).
Note that Z|Xn X| 1 a.s. and Xn X a.s. by hypothesis, hence by dominated
convergence theorem EQ (|Xn X|) = EP (Z|Xn X|) 0, i.e. Xn tends to X in L1 with
respect to Q.
17.14 First observe that |E(Xn2 ) E(X 2 )| E(|Xn2 X 2 |). Since |Xn2 X 2 | (Xn X)2 +
2|X||Xn X| we get that |E(Xn2 ) E(X 2 )| E((Xn X)2 ) + 2E(|X||Xn X|). Note
that first term goes to 0 since Xn tends to X in p
L2 . Applying Cauchy Schwarz inequality
to the second term we get E(|X||Xn X|) E(X 2 )E(|Xn X|2 ), hence the second
term also goes to 0 as n . Now we can conclude E(Xn2 ) E(X 2 ).
17.15 For any  > 0 P ({|X| c+}) P ({|Xn | c, |Xn X| }) 1 as n . Hence
1
P ({|X| c + }) = 1. Since {X c} =
m=1 {X c + m } we get that P {X c} = 1.
Now we have that E(|Xn X|) = E(|Xn X|1{|Xn X|} ) + E(|Xn X|1{|Xn X|>} )
 + 2c(P {|Xn X| > }), hence choosing n large we can make E(|Xn X|) arbitrarily
small, so Xn tends to X in L1 .

18

Solutions to selected problems of Chapter 18

|u|

18.8 Note that Yn (u) = ni=1 Xi ( nu ) = ni=1 e n = e|u| , hence Yn is also Cauchy with
= 0 and = 1 which is independent of n, hence trivially Yn converges in distribution
to a Cauchy distributed r.v. with = 0 and = 1. However Yn does not converge to
any r.v. in probability. To see this, suppose there exists Y s.t. P (|Yn Y | > ) 0.
Note that P (|Yn Ym | > ) P (|Yn Y | > 2 ) + P (|Ym Y | > 2 ). If we let m = 2n,
P
P
1
|Yn Ym | = 12 | n1 ni=1 Xi n1 2n
i=n+1 Xi | which is equal in distribution to 2 |U W | where
U and W are independent Cauchy r.v.s with = 0 and = 1. Hence P (|Yn Ym | > 2 )
does not depend on n and does not converge to 0 if we let m = 2n and n which is a
contradiction since we assumed the right hand side converges to 0.
18.16 Define fm as the following sequence of functions:

x2

(N m1 )x (N m1 )N
fm (x) =
(N m1 )x + (N m1 )N

if |x| N m1
if x N m1
if x N + m1
otherwise

Note that each fm is continuous and bounded. Also fm (x) 1(N,N ) (x)x2 for every x R.
Hence
Z N
Z
2
x F (dx) = lim
fm (x)F (dx)
m

by monotone convergence theorem. Now

Z
Z
fm (x)F (dx) = lim
n

fm (x)Fn (dx)

RN
R
by weak convergence. Since fm (x)Fn (dx) N x2 Fn (dx) it follows that
Z N
Z N
Z N
2
2
x F (dx) lim lim sup
x Fn (dx) = lim sup
x2 Fn (dx)
N

as desired.
18.17 Following the hint, suppose there exists a continuity point y of F such that
lim Fn (y) 6= F (y)

Then there exist  > 0 and a subsequence (nk )k1 s.t. Fnk (y) F (y) <  for all k, or
Fnk (y) F (y) >  for all k. Suppose Fnk (y) F (y) <  for all k, observe that for x y,
Fnk (x) F (x) Fnk (y) F (x) = Fnk (y) F (y) + (F (y) F (x)) <  + (F (y) F (x)).
Since f is continuous at y there exists an interval [y1 , y) s.t. |(F (y) F (x))| < 2 , hence
Fnk (x) F (x) < 2 for all x [y1 , y). Now suppose Fnk (y) F (y) > , then for x y,
Fnk (x) F (x) Fnk (y) F (x) = Fnk (y) F (y) + (F (y) F (x)) >  + (F (y) F (x)).
19

Now we can find an interval (y, y1 ] s.t. |(F (y) F (x))| < 2 which gives Fnk (x) F (x) >
for all x (y, y1 ]. Note that both cases would yield
Z

|Fnk (x) F (x)|r dx > |y1 y|
2
inf ty
which is a contradiction to the assumption
Z
|Fn (x) F (x)|r dx = 0.
lim
n

inf ty

Therefore Xn converges to X in distribution.

20


2

Solutions to selected problems of Chapter 19

2
u2 n

19.1 Note that Xn (u) = eiun 2 eiu

that Xn X where X is N (, 2 ).

u2 2
2

. By Levys continuity theorem it follows

19.3 Note that Xn +Yn (u) = Xn (u)Yn (u) X (u)Y (u) = X+Y (u). Therefore Xn +
Yn X + Y

21

Solutions to selected problems of Chapter 20

P P
P
20.1 a. First observe that E(Sn2 ) = ni=1 nj=1 E(Xi Xj ) = ni=1 Xi2 since E(Xi Xj ) = 0
2

nE(X 2 )

n)
for i 6= j. Now P ( |Snn | ) E(S
= 2 n2i nc 2 as desired.
2 n2
b. From part (a) it is clear that n1 Sn converges to 0 in probability. Also E(( n1 Sn )2 ) =

E(Xi2
n

0 since E(Xi2 ) , so n1 Sn converges to 0 in L2 as well.

20.5 Note that Zn Z implies that Zn (u) Z (u) uniformly on compact subset of R.
(See Remark 19.1). For any u, we can pick n > N s.t. un < M , supx[M,M ] |Zn (x)
Z (x)| <  and |varphiZ ( un ) Z (0)| < . This gives us
u
u
u
u
|Zn ( ) Z (0)| = |Zn ( ) Z ( )| + |Z ( ) Z (0)| 2
n
n
n
n
So Zn (u) = Zn ( un ) converges to Z (0) = 1 for every u. Therefore
n

Zn

0 by continuity

Zn
theorem. We also have by the strong law of large numbers that
E(Xj ) . This
n
implies E(Xj ) = 0, hence the assertion follows by strong law of large numbers.

22

http://www.springer.com/978-3-540-43871-7