Central Limit Theorem
It is possible to illustrate the Central Limit Theorem empirically. It is also possible to illustrate the Central Limit Theorem theoretically. The following examples show how this can be done.
1. [Exercise 5.4.4 and Figure 5.5 in Hogg/Tanis] Let
, ...,
be a random sample of size
n
from a
U
-shaped distribution with p.d.f.
f(x)
=
,
<
x
< 1. Empirically show that
is approximately
N
(0, 1) when
n
is sufficiently large.
Solution
2. [Exercise 5.4.10 and Figure 5.5 in Hogg/Tanis] Let
, ...,
be a random sample of size
n
from a
U
-shaped distribution with p.d.f.
,
<
x
< 1. Theoretically show that the p.d.f. of
=
becomes closer and closer to the N(0,1) p.d.f. as
n
increases.
Solution
3. [Exercise 5.4.11(a)] Let
, ...,
be a random sample of size
n
from the uniform distribution on the interval [
, 1]. I.e., the distribution is
U
(
, 1). Show that the p.d.f. of
=
becomes closer and closer to the
N
(0, 1) p.d.f. as
n
increases.
4. [Exercise 5.4.11(b) and Figure 5.4 of Hogg/Tanis] Let
, ...,
be a random sample of size
n
from a triangular distribution on the interval [
, 1] with p.d.f. defined by f(x) =
,
< x < 1. For this distribution,
,
. Show that the p.d.f. of
=
becomes closer and closer to the
N
(0, 1) p.d.f. as
n
increases.