1

LIMITING DISTRIBUTIONS
2
CONVERGENCE IN DISTRIBUTION
Consider that X
1
, X
2
,, X
n
is a sequence of
rvs and Y
n
=u(X
1
, X
2
,, X
n
) be a function of
rvs with cdfs F
n
(y) so that for eachn=1, 2,
( ) ( )
,
n n
F y P Y y =
( ) ( )
lim for all
n
n
F y F y y

=
where F(y) is continuous. Then, the sequence
X
1
, X
2
,, X
n
is said to converge in distribution.
d
n
Y Y
3
CONVERGENCE IN DISTRIBUTION
Theorem: If for every
point y at which F(y) is continuous, then Y
n
is said to have a limiting distribution with
cdf F(y).
Definition of convergence in distribution
requires only that limiting function agrees
with cdf at its points of continuity.
( ) ( )
lim
n
n
F y F y

=
4
EXAMPLES
1. Let {X
n
} be a sequence of rvs with pmf
( ) ( )
1
1, if 2
0, o.w.
n
x
f x P X x
n

= +

= = =

Find the limiting distribution of X

n
.
5
EXAMPLES
2. Let X
n
have the pmf
( )
1 if .
n
f x x n = =
Find the limiting distribution of X
n
.
6
EXAMPLES
3. Let X
n
~ N(,
2
) be a sequence of Normal
rvs. Let be the sample mean. Find the
limiting distribution of .
n
X
n
X
7
CONVERGENCE IN PROBABILITY
(STOCHASTIC CONVERGENCE)
A rv Y
n
convergence in probability to a rv Y if
( )
lim 1
n
n
P Y Y

< =
for every >0.
Special case: Y=c where c is a constant
not depending on n.
The limiting distribution of Y
n
is
degenerate at point c.
8
CHEBYSHEVS INEQUALITY
Let X be an rv with E(X)= and V(X)=
2
.
( )
2
2
1 , 0 P X < >

The Chebyshevs Inequality can be

used to prove stochastic convergence in
many cases.
9
CONVERGENCE IN PROBABILITY
(STOCHASTIC CONVERGENCE)
The Chebyshevs Inequality proves the
convergence in probability if the following
three conditions are satisfied.
( )
2
2. for all .
n n
V Y n = <
1. E(Y
n
)=
n
where
lim .
n
n
=
2
3. lim 0.
n
n
=
10
EXAMPLES
1. Let X be an rv with E(X)= and V(X)=
2
<.
For a r.s. of size n, is the sample mean.
Is
n
X
?
p
n
X
11
EXAMPLES
Let Z
n
be
2 2
and let / .
n n n
W Z n =
Show that the limiting distribution of W
n
is
degenerate at 0.
12
WEAK LAW OF LARGE
NUMBERS
Let X
1
, X
2
,,X
n
be iid rvs with E(X
i
)= and
V(X
i
)=
2
<. Define . Then, for
every >0,
1
1/
n
n i
i
X n X
=
=

( )
lim 1,
n
n
P X

< =
that is, converges in probability to .
n
X
13
STRONG LAW OF LARGE
NUMBERS
Let X
1
, X
2
,,X
n
be iid rvs with E(X
i
)= and
V(X
i
)=
2
<. Define . Then, for
every >0,
1
1/
n
n i
i
X n X
=
=

( )
lim 1
n
n
P X

< =
that is, converges almost sure to .
n
X
14
LIMITING MOMENT
GENERATING FUNCTIONS
Let rv Y
n
have an mgf M
n
(t) that exists for
all n. If
( ) ( )
lim ,
n
n
M t M t

=
then Y
n
has a limiting distribution which is
defined by M(t).
15
EXAMPLES
1. Let X
n
~ Gamma(n, ) where does not
depend on n. Let Y
n
=X
n
/n. Find the limiting
distribution of Y
n
.
16
EXAMPLES
2. Let X
n
~ Exp(1) and be the sample
mean of r.s. of size n. Find the limiting
distribution of
n
X
( )
1 .
n n
Y n X =
17
THE CENTRAL LIMIT THEOREM
Let X
1
, X
2
,,X
n
be a sequence of iid rvs
whose mgf exist in a neighborhood of 0.
Let E(X
i
)= and V(X
i
)=
2
>0. Define
. Then,
1
1/
n
n i
i
X n X
=
=

( )
( )
0,1
d
n
n X
Z N

or
( )
1
0,1 .
n
i
d
i
X n
Z N
n
=

=

18
EXAMPLES
1. Let X
n
~ Exp(1) and be the sample
mean of r.s. of size n. Find the limiting
distribution of
n
X
( )
1 .
n n
Y n X =
19
EXAMPLES
2. Let be the sample mean from a r.s. of
size n=100 from . Compute approximate
value of
n
X
2
50

( )
49 51 . P X < <
20
SLUTKYS THEOREM
If X
n
X in distribution and Y
n
a, a
constant, in probability, then
a) Y
n
X
n
aX in distribution.
b) X
n
+Y
n
X+a in distribution.
21
SOME THEOREMS ON LIMITING
DISTRIBUTIONS
If X
n
c>0 in probability,
.
p
n
X c
If X
n
c in probability and Y
n
c in probability,
then
aX
n
+bY
n
ac+bd in probability.
X
n
Y
n
cd in probability
1/X
n
1/c in probability for all c0.
22
EXAMPLES
1. X~Gamma(, 1). Show that
( )
( )
0,1.
d
n
n
n X
N
X

23
EXAMPLES
X~Gamma(1,n). Let
n
n
X n
Z
n

=
Let Z
n
N(0,1) in distribution and Y
n
c in
probability. Find the limiting distribution
of the following
a)W
n
=Y
n
Z
n
.
b)U
n
=Z
n
/n.
c)V
n
=Z
n
+Y
n
.
24
ORDER STATISTICS
Let X
1
, X
2
,,X
n
be a r.s. of size n from a
distribution of continuous type having pdf
f(x), a<x<b. Let X
(1)
be the smallest of X
i
,
X
(2)
be the second smallest of X
i
,, and
X
(n)
be the largest of X
i
.
( ) ( ) ( ) 1 2 n
a X X X b < < L
X
(i)
is the i-th order statistic.
( )
{ }
( )
{ }
1 2 1
1 2
min , , ,
max , , ,
n
n n
X X X X
X X X X
=
=
L
L
25
ORDER STATISTICS
If X
1
, X
2
,,X
n
be a r.s. of size n from a
population with continuous pdf f(x), then
the joint pdf of the order statistics
X
(1)
, X
(2)
,,X
(n)
is
( ) ( ) ( )
( )
( )
( )
( )
( )
( )
( )
1 2 1 2
, , , !
n n
g x x x n f x f x f x = L L
26
ORDER STATISTICS
The Maximum Order Statistic: X
(n)
( )
( )
( )
( )
n
X
n
G y P X y =
( )
( )
( )
( )
n n
X X
g y G y
y

27
ORDER STATISTICS
The Minimum Order Statistic: X
(1)
( )
( )
( )
( )
1
1
X
G y P X y =
( )
( )
( )
( )
1 1
X X
g y G y
y

28
ORDER STATISTICS
k-th Order Statistic
y
y
1
y
2
y
k-1
y
k
y
k+1
y
n

P(X<y
k
) P(X>y
k
)
f
X
(y
k
)
# of possible orderings
n!/{(k1)!1!(n k)!}
( )
( )
( ) ( )
( ) ( ) ( )
1
!
1 ,
1! !
k
k n k
X X X
X y
n
g F y f y F y a y b
k n k

= < <


29
EXAMPLE
X~Uniform(0,). A r.s. of size n is taken. X
(n)
is the largest order statistic. Then,
a) Find the limiting distribution of X
(n).
b) Find the limiting distribution of Z
n
=n(X
(n)
).