Vous êtes sur la page 1sur 1

Theorem The beta(b, b) distribution converges to the normal distribution when b → ∞.

Proof (by Professor Robin Ryder in the CEREMADE at Université Paris Dauphine) Let
the random variable X have the beta(b, b) distribution with probability density function

Γ(2b)xb−1 (1 − x)b−1
fX (x) = 0 < x < 1,
Γ(b)Γ(b)
where b is a real, positive parameter. The mean of X is E[X] = 1/2 and the variance of
X is V [X] = 1/4(2b + 1). Substract the mean and divide by the
√ standard deviation before
taking the limit. So consider the transformation Y = g(X) = 2 2b√+ 1(X − 1/2),√which is
a one-to-one transformation from√A = {x|0 < x < 1} to B = {y| − 2b + 1 < y < 2b + 1}
with inverse X = g −1 (Y ) = X/2 2b + 1 + 1/2 and Jacobian
dX 1
= √ .
dY 2 2b + 1
Then the probability density function of Y is
 
1 y 1
fY (y) = √ fX √ +
2 2b + 1 2 2b + 1 2
 b−1  b−1
1 Γ(2b) 1 y 1 y
= √ · + √ − √
2 2b + 1 Γ(b)2 2 2 2b + 1 2 2 2b + 1
b−1
y2 √ √

1 Γ(2b) 1
= √ · − − 2b + 1 < y < 2b + 1.
2 2b + 1 Γ(b)2 4 4(2b + 1)
p
We now apply Stirling’s formula Γ(z) = 2π/z(z/e)z (1 + O(1/z)) and get
q
2π 2b 2b 
 b−1
y2
  
1 2b e 1 1
fY (y) = √ · 1+O −
2 2b + 1 2π b 2b b 4 4(2b + 1)

b e
2b
b−1
y2
   
b 2 1 1
= √ √ · √ 1+O −
2b 2b + 1 2 2π b 4 4(2b + 1)
b−1
y2
   
1 b−1 1 1
= √ 4 1+O −
2π b 4 4(2b + 1)
b−1 
y2
  
1 1
= √ 1− 1+O
2π 2b + 1 b
√ √
 2  
1 y 1
= √ exp − 1+O − 2b + 1 < y < 2b + 1.
2π 2 b
Thus, in the limit b → ∞, fY (y) converges pointwise to the probability density function of a
standard normal random variable. By Scheffé’s theorem, Y converges in distribution to the
standard normal distribution.