Ftshort | R Documentation |
Calculate values of 1 - 4-term Edgeworth expansions (EE) (2nd - 5th order) for ordinary one-sample t-statistic (short version).
Ft1(x, n, lam, r = sqrt((n - 1)/n), norm = TRUE, df = n - 1)
Ft2(x, n, lam, r = sqrt((n - 1)/n), norm = TRUE, df = n - 1)
Ft3(x, n, lam, r = sqrt((n - 1)/n), norm = TRUE, df = n - 1)
Ft4(x, n, lam, r = sqrt((n - 1)/n), norm = TRUE, df = n - 1)
x |
numeric vector of quantiles of sampling distribution. |
n |
sample size. |
lam |
a vector of scaled cumulants or their estimates, starting with 3rd (skewness). Term 1 needs a third cumulant only, term 2 - third and fourth (kurtosis), and so on. |
r |
square root of variance adjustment (see "Details"). |
norm |
if |
df |
degrees of freedom for Student's t-distribution if |
Higher-order approximations of the cumulative distribution function of a
t-statistic. These functions implement a short version of EE that can be used
for ordinary one-sample t-statistic. Note that in this case the variance
adjustment r^2
is equal to 1 for a t-statistic with naive biased
variance estimate and (n - 1)/n
for a standard t with unbiased variance
estimate.
A vector of the same length as x
containing the values of
Edgeworth expansion of a corresponding order (Ft1
for a 1-term or
2nd order EE, Ft2
for a 2-term EE, and so on).
qfuns
for q()
functions used in EE terms and
Ftgen
for a general version of EE. For creating EE as a
simple function of x
, see makeFx
.
# Gamma distribution with shape parameter shp
n <- 10
shp <- 3
ord <- 3:6 # orders of scaled cumulants
lambdas <- factorial(ord - 1)/shp^((ord - 2)/2)
x <- seq(-5, -2, by = 0.5) # thicker tail
Ft1(x, n, lambdas, r = 1)
Ft1(x, n, lambdas, norm = FALSE)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.