TY - JOUR
T1 - Variations on the berry-esseen theorem
AU - Klartag, B.
AU - Sodin, S.
PY - 2012
Y1 - 2012
N2 - Suppose that X 1,...,X n are independent, identically distributed random variables of mean zero and variance one. Assume that E|X 1| 4 ≦ δ 4. We observe that there exist many choices of coefficients θ 1,..., θ n ∈ R with Σ j θ 2 j = 1 for which sup, where C > 0 is a universal constant. This inequality should be compared with the classical Berry-Esseen theorem, according to which the left-hand side may decay with n at the slower rate of O(1/ √ n) for the unit vector θ = (1, ..., 1)/ √ n. An explicit, universal example for coefficients θ = (θ1, ..., θn) for which this inequality holds is θ = (1, √ 2,-1,- √ 2, 1, √ 2,-1,- √ 2, ...)(3n/2) -1/2, when n is divisible by four. Parts of the argument are applicable also in the more general case, in which X 1, ..., X n are independent random variables of mean zero and variance one yet are not necessarily identically distributed. In this general setting, the bound above holds with δ 4 = n -1Σ n j=1 E|X j | 4 for most selections of a unit vector θ = (θ 1, ..., θ n) ∈ R n. Here "most" refers to the uniform probability measure on the unit sphere.
AB - Suppose that X 1,...,X n are independent, identically distributed random variables of mean zero and variance one. Assume that E|X 1| 4 ≦ δ 4. We observe that there exist many choices of coefficients θ 1,..., θ n ∈ R with Σ j θ 2 j = 1 for which sup, where C > 0 is a universal constant. This inequality should be compared with the classical Berry-Esseen theorem, according to which the left-hand side may decay with n at the slower rate of O(1/ √ n) for the unit vector θ = (1, ..., 1)/ √ n. An explicit, universal example for coefficients θ = (θ1, ..., θn) for which this inequality holds is θ = (1, √ 2,-1,- √ 2, 1, √ 2,-1,- √ 2, ...)(3n/2) -1/2, when n is divisible by four. Parts of the argument are applicable also in the more general case, in which X 1, ..., X n are independent random variables of mean zero and variance one yet are not necessarily identically distributed. In this general setting, the bound above holds with δ 4 = n -1Σ n j=1 E|X j | 4 for most selections of a unit vector θ = (θ 1, ..., θ n) ∈ R n. Here "most" refers to the uniform probability measure on the unit sphere.
KW - Berry-esseen theorem
KW - Central limit theorem
KW - Gaussian distribution
UR - http://www.scopus.com/inward/record.url?scp=84867726378&partnerID=8YFLogxK
U2 - 10.1137/S0040585X97985522
DO - 10.1137/S0040585X97985522
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:84867726378
SN - 0040-585X
VL - 56
SP - 403
EP - 419
JO - Theory of Probability and its Applications
JF - Theory of Probability and its Applications
IS - 3
ER -