The bias-adjusted sample excess kurtosis, a single number;
NA when degenerate (see Details).
Arguments
x
A numeric vector.
na.rm
Logical; remove NA values before computing. Default
TRUE.
Details
The bias-adjusted Fisher-Pearson standardized EXCESS kurtosis
coefficient, G2 = ((n + 1) * g2 + 6) * (n - 1) / ((n - 2) * (n -
3)), where g2 = m4 / m2^2 - 3 and m2/m4 are the
second/fourth central sample moments of x -- the "Method 2"
adjustment of Joanes and Gill (1998), the same convention SPSS, SAS, and
Excel report by default and the type = 2 option in the
e1071 CRAN package. Excess (not raw) kurtosis: a
normal distribution reads 0. Positive values indicate heavier
tails / a sharper peak than normal; negative, lighter tails / a flatter
peak.
Returns NA when na.rm = FALSE and x contains an
NA, when x has fewer than 4 non-NA values
(n <= 3, the adjustment term divides by (n - 2) * (n - 3))
or has zero variance (all remaining values identical) -- both would
otherwise divide by zero.
References
Joanes, D.N. and Gill, C.A. (1998) "Comparing measures of
sample skewness and kurtosis." Journal of the Royal Statistical
Society: Series D (The Statistician), 47(1), 183-189.
See Also
calcSkewness
Other genetic value analysis:
calcA(),
calcFE(),
calcFEFG(),
calcFG(),
calcFGSE(),
calcGU(),
calcGUSE(),
calcGeneDiversity(),
calcNeSexRatio(),
calcNeVariance(),
calcRetention(),
calcSkewness()