ORTHOGONAL POLYNOMIALS;
CENTRAL LIMIT THEOREM;
FOURIER TRANSFORM;
D O I:
10.1016/0377-0427(93)E0233-C
中图分类号:
O29 [应用数学];
学科分类号:
070104 ;
摘要:
We study the asymptotic behaviour of normalized sums of n random variables Sigma(n) = n (-1/2)(X(1) + X(2) + ... + X(n)). We obtain a necessary and sufficient condition in order that the moments of Sigma(n) converge to the moments of a normal distribution. This condition is expressed by means of polynomial dependence coefficients rho(j1,j2) (...) (,jn). Then, one can obtain central limit theorems with minimal assumptions.