Scribd
Upload
19 views

Central Limit Theorem

(1) The document introduces the Central Limit Theorem (CLT), which states that for large sample sizes, the sum of independent random variables with finite mean and variance will be approximately normally distributed, regardless of the underlying distribution. (2) Specifically, the CLT says that the distribution of the sample mean will approach a normal distribution, and the distribution of the standardized sample mean will approach the standard normal distribution. (3) As a special case, the document notes that the binomial distribution can be viewed as a sum of random variables and thus approximated by the normal distribution for large samples via the CLT.

Uploaded by

guipingxie
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
19 views

Central Limit Theorem

(1) The document introduces the Central Limit Theorem (CLT), which states that for large sample sizes, the sum of independent random variables with finite mean and variance will be approximately normally distributed, regardless of the underlying distribution. (2) Specifically, the CLT says that the distribution of the sample mean will approach a normal distribution, and the distribution of the standardized sample mean will approach the standard normal distribution. (3) As a special case, the document notes that the binomial distribution can be viewed as a sum of random variables and thus approximated by the normal distribution for large samples via the CLT.

Uploaded by

guipingxie
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Math 461

Introduction to Probability

A.J. Hildebrand

The Central Limit Theorem


Setup: Let X1 , X2 , X3 . . . be independent, identically distributed (iid) random variables, i.e.,
r.v.s that (i) are mutually independent, and (ii) have the same distribution with finite expectation and
variance given by
2 = Var(Xi ),

= E(Xi ),
Let
Sn =

n
X

Xi .

i=1

Central Limit Theorem (CLT):


Normal approximation form: For large n,
Sn is approximately normal N (n, n 2 )
or, equivalently,
Sn n

is approximately standard normal N (0, 1)


n
Limit form: For all z,

lim P

Sn n

z
n


= (z).

Special case of normal r.v.s: In the special case when the Xi s are already normal with
mean and 2 , we know that the exact distribution of Sn is N (n, n 2 ), by the formula for the
distribution of a sum of independent normals.
Thus, the CLT can be interpreted as saying that, when n is large, then a sum of n independent
r.v.s with an arbitrary common distribution behaves as if the individual variables were normal with
the same mean and variance.
Normal approximation to the binomial distribution: Another type of normal approximation
came up in Chapter 5: A binomial distribution with parameters n and p is approximately normal
N (np, np(1 p)) provided n is large and p is not too close to 0 or to 1. This result can be obtained
as a special
case of the CLT, by representing the number of successes in n S/F trials as a sum
P
Sn = ni=1 Xi , where Xi is the indicator random variable of the event success at trial i, and has
mean = E(Xi ) = p and variance 2 = Var(Xi ) = p(1 p).
Weak Law of Large Numbers: For any  > 0,




Sn


 = 1.
lim P
n
n
Probabilistic inequalities:
Markov inequality:
a > 0,

Let X be a nonnegative (i.e., X 0) r.v. with mean . Then, for any

.
a
Chebychev inequality: Let X be a r.v. with mean and variance 2 . Then, for any k > 0,
P (X a)

P (|X | k)

2
.
k2

You might also like