4.3. THE CENTRAL LIMIT THEOREM 137
wherer 2 (t/n) is a error term such that
lim
n→∞
r(t/n)
(1/n)
= 0.
Then we need to consider
φ
(
t
n
)n
= (1 +μ
t
n
+r 2 (t/n))n.
Taking the logarithm of (1 +μ(t/n) +r(t/n))nand using L’Hospital’s Rule,
we see that
φ(t/n)n→exp(μt).
But this last expression is the moment generating function of the (degenerate)
point mass distribution concentrated atμ. Hence,
lim
n→∞
P[|Sn/n−μ|> ] = 0
The Central Limit Theorem
Theorem 14(Central Limit Theorem).Let random variablesX 1 ,...Xn
- be independent and identically distributed,
- have common meanE[Xi] =μand common varianceVar [Xi] =σ^2 ,
- the common moment generating functionφXi(t) =E[etxi]exists and is
finite in a neighborhood oft= 0.
ConsiderSn=
∑n
i=1Xi.Let
- Zn= (Sn−nμ)/(σ
√
n) = (1/σ)(Sn/n−μ)
√
n,
- Z be the standard normally distributed random variable with mean 0
and variance 1.
ThenZnconverges in distribution toZ, that is:
lim
n→∞
P[Zn≤a] =
∫a
−∞
(1/
√
2 π) exp(−u^2 /2)du.