Math 408, Actuarial Statistics I

A.J. Hildebrand

The Central Limit Theorem

Random samples, iid random variables

Denition: A random sample of size n from a given distribution is a set of n independent r.v.s X1 , X2 , . . . , Xn , each having the given distribution, with expectation E (Xi ) = and variance Var(Xi ) = 2 . Such a set of random variables is also called independent, identically distributed (iid). Sample sum: S = Sample mean: X =
n i=1 Xi . 1 n

E (S ) = n, Var(S ) = n 2 , E (X ) = , Var(X ) = 2 /n

n i=1 Xi .

Normal approximation, Central Limit Theorem

The Central Limit Theorem (CLT) says that the mean and the sum of a random sample of a large enough size1 from an (essentially) arbitrary distribution have approximately normal distribution: Given a random sample X1 , . . . , Xn with = E (Xi ) and 2 = Var(Xi ), we have: The sample sum S = is approximately normal N (n, n 2 ). S n , is approximately standard Equivalently, the standardized version of S , S = n normal. is approximately normal N (, 2 /n). X , is approximately standard Equivalently, the standardized version of X , X = / n normal.
1 n n i=1 Xi n i=1 Xi

The sample mean X =

Sums of independent normal r.v.s

In the case the Xi s are already normal, the associated sums and random samples are exactly (and not just approximately) normal, with the appropriate parameters:
2 ) and Sums and dierences of two normal r.v.s: If X1 and X2 are N (1 , 1 2 ), respectively, then X X is normal N ( , 2 + 2 ). (Note the plus N (2 , 2 1 2 1 2 1 2 sign in the formula for the variance of X1 + X2 . Also, note that it is the variances that add up, not the standard deviations.) 2 ), then General linear combinations of normal r.v.s: If each Xi is normal N (i , i n n n 2 2 2 2 i=1 ci Xi is normal N (, ), where = i=1 ci i and = i=1 ci i .

That is, a large enough value of n. The larger n is, the better the CLT approximation becomes, but greater than 25 is usually more than enough in practice, and in some cases the CLT works already well for single digit values of n.