robustloggamma (version 1.0-2.1)

summary.loggammarob: Summary Method for "loggammarob" Objects

Description

Summary method for R object of class "loggammarob" and print method for the summary object.

Usage

# S3 method for loggammarob
summary(object, p = NULL, conf.level = 0.95,
  prob = 1e-05, ...)
# S3 method for summary.loggammarob
print(x, digits = max(3, getOption("digits") - 3), ...)

Arguments

object

an R object of class loggammarob, typically created by loggammarob.

p

numeric. Values in the interval [0,1]. Quantile orders for which point estimation and confidence interval are require.

conf.level

numeric. A scalar or a vector of length 4+length(p). This is the confidence level used to construct confidence intervals for the four parameters mu, sigma, lambda and eta and for the quantiles.

prob

numeric. Value in the interval [0,1]. This is used to determine the interval of numerical integration in the evaluation of the asymptotic variance and covariance matrix. See details below.

x

an R object of class summary.loggammarob, typically resulting from summary(loggammarob(..),..).

digits

number of digits for printing, see digits in options.

…

potentially more arguments passed to methods.

Value

An object of class summary.loggammarob. A list that includes the following components:

muse

standard error for the mu estimate.

sigmase

standard error for the sigma estimate.

lambdase

standard error for the lambda estimate.

etase

standard error for the eta estimate.

muconf.int

confidence iterval for the mu parameter.

sigmaconf.int

confidence iterval for the sigma parameter.

lambdaconf.int

confidence iterval for the lambda parameter.

etaconf.int

confidence iterval for the eta parameter.

If p is not NULL then

q

quantiles estimate.

qse

standard error for the require quantiles estimate.

qconf.int

confidence iterval for the require quantiles parameter.

Details

The prob argument determines the interval of the numerical integration in the evaluation of the asymptotic variance and covariance matrix with the following code qloggamma(p=prob/2,lambda=lambda) for the lower limit and qloggamma(p=1-prob/2,lambda=lambda) for the upper limit. Here lambda is the estimate.

References

C. Agostinelli, A. Marazzi and V.J. Yohai (2015). Robust estimates of the generalized loggamma distribution. Technometrics, Volume 56, Issue 1, 2014. doi:10.1080/00401706.2013.818578

Agostinelli C., Marazzi A., Yohai V.J., Randriamiharisoa A. (2016). Robust Estimation of the Generalized Loggamma Model: The R Package robustloggamma. Journal of Statistical Software, 70(7), 1-21. doi:10.18637/jss.v070.i07

See Also

loggammarob

Examples

# NOT RUN {
  set.seed(1234)
  x <- rloggamma(n=50, lambda=1)
  res <- loggammarob(x, control=loggammarob.control(lower=0, upper=2, n=30))
  summary(res)
# }