Edgeworth series

E953116

The Edgeworth series is a statistical expansion used in probability theory to approximate probability distributions by correcting the normal distribution with higher-order cumulants.

Jump to: Statements Referenced by

Statements (44)

Predicate Object
instanceOf asymptotic expansion
probability theory concept
statistical method
appliesTo independent and identically distributed random variables
sums of random variables
approximationOrder can be extended to arbitrary finite order in 1/sqrt(n)
assumes existence of sufficiently high-order moments
basedOn cumulants
moments of a distribution
comparedWith Gram–Charlier A series NERFINISHED
domain continuous distributions
discrete distributions (via continuity corrections)
expandsAround normal distribution
field mathematical statistics
probability theory
generalizationOf normal approximation
goal to approximate distribution functions more accurately than the normal distribution alone
hasAdvantageOver Gram–Charlier series in asymptotic justification
hasProperty asymptotic in sample size
may not define a proper probability density for finite truncation
historicalPublicationPeriod late 19th century
improves rate of convergence of normal approximation
namedAfter Francis Ysidro Edgeworth NERFINISHED
relatedTo Gram–Charlier series NERFINISHED
central limit theorem NERFINISHED
saddlepoint approximation
representation polynomial corrections to the normal density
series in powers of n^{-1/2}
requires standardization of the random variable
usedBy applied probabilists
econometricians
statisticians
usedFor approximating distribution of standardized sums
approximating probability distributions
approximating sampling distributions
refining normal approximation
usedIn confidence interval approximation
econometrics
hypothesis testing
theoretical statistics
usesConcept higher-order cumulants
kurtosis
skewness
standardized cumulants

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

Francis Ysidro Edgeworth knownFor Edgeworth series